| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Departments of
*
Surgery and
Pathology and
Transplantation Biology Research Center, Harvard Medical School, Massachusetts General Hospital, Boston, MA 02114
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
It might be possible to control this immune response more completely by higher doses of available immunosuppressive agents or perhaps by new agents targeted to selected features of the inflammatory response, such as to certain cytokines, as suggested by the interesting recent experiments of Hancock et al. (3). Still, the possibility of eliminating these arteriopathic lesions and the accompanying changes of chronic rejection by inducing some form of high-grade, specific unresponsiveness to the appropriate Ags of the transplanted organ has remained an attractive one. Should this be achievable, such a state of nonreactivity, or immunological tolerance, would ideally result in the complete and long-lasting elimination of the specific immune response to all the cells of the allospecific donor tissue and should thereby eliminate all of the effects of chronic rejection and the need for continuing immunosuppression with its undesirable side effects.
This report is to describe the present status of our experiments to evaluate the efficacy of two forms of specific nonreactivity, which can be reliably induced and verified under standardized immunogenetic conditions, on the development of arteriopathic lesions in hearts transplanted between mice. As an extension of these experiments, we have also evaluated the outcome of hearts transplanted to mice nonspecifically incapable of immune responses following targeted genetic alterations.
The first example of specific nonreactivity we have selected has been "actively acquired immunological tolerance" induced by the i.v. administration shortly after birth of lymphoid cells from adult members of an allogeneic strain. This we have done exactly according to the original scheme of Billingham et al. (4) and by the technique subsequently described by Billingham and Brent (5). The second form of specific unresponsiveness has been created in adult recipients by the production of "mixed chimerism." This is achieved by delivering bone marrow cells from the donor strain to adult recipients that have been prepared with a nonmyeloablative regimen that leaves recipient hemopoietic cells intact. The methods used have been studied extensively by members of our group and have undergone steady improvement, particularly in the measures required to prepare recipients for infusions of allogeneic cells (6, 7, 8). In both situations, the experiments have included the same strain combination of animals. The status of cell recipients and the outcomes of treatment in both have also been compared using the same methods.
In additional experiments in which allogeneic heart transplants were performed on animals with different forms of profound nonspecific immune deficiencies, no additional treatment of recipients was given. Finally, isologous heart transplants were also performed as contemporary controls beyond our previous experience already reported (9).
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
After the induction of specific tolerance by either of the methods under study, the nonreactivity of each treated recipient was first verified in most experiments by the application of a test skin graft from the donor strain. The strain combination selected was B10.A donors to B10.BR recipients (H-2a to H-2k, a class I Ag difference at H-2D). This combination was used for both approaches to tolerance. In addition, B10.A to C57BL/6 (H-2a to H-2b, involving a full disparity of all class I and II Ags) was included in studies of tolerance by mixed chimerism. The immunological status of treated recipients was also examined further by evaluating the existence and level of persisting donor cell chimerism from surviving descendents of the originally delivered cells. Information was also derived about the presence and type of chimeric cells within treated recipients as well as the presence of detectable levels of humoral Abs to donor-specific Ags. Finally, using such tolerant animals as further recipients, heterotopic heart transplants from the donor strains were performed. Some mixed chimeric recipients did not undergo skin grafting before receiving heart transplants. Heart transplants were allowed to survive for extended periods before termination of the experiments, at which time their coronary vessels were evaluated histologically.
In experiments using recipients with profound, but nonspecific, immunological unresponsiveness, mice with SCID, C.B-17/SCID (H-2d), C.B-17/SCID/beige (H-2d), and those with a recombinant-activating gene deficiency, RAG1-deficient (H-2b), received heart transplants from B10.BR strain donors.
Mice
B10.A, B10.BR, C57BL/6, and RAG1-/- mice were all obtained from The Jackson Laboratory (Bar Harbor, ME). C.B-17/SCID/beige mice were obtained from Taconic Farms (Germantown, NY) and C.B-17/SCID mice were obtained from the breeding colony of the Department of Radiation Medicine of our institution. RAG1-/- and SCID mice were maintained in filter top cages and remained entirely healthy throughout the experiments. In selected RAG1-/- mice, as detailed below, their bona fide designation as RAG1-/- animals was confirmed by appropriate DNA isolation for the presence of the inserted neomycin resistance gene and the absence of the recombination activation gene-1 gene as described previously (10). F1 generation mice between the B10.A and B10.BR strains were bred in our laboratory. All animals were cared for according to American Association for the Accreditation of Laboratory Animal Care approved methods. For the mixed chimerism experiments 8- to 12-wk-old female C57BL/6 (H-2b) and B10.A (H-2a) mice were purchased from the Frederick Cancer Research Center (Frederick, MD). In one experiment, mice were also maintained throughout in specific pathogen-free conditions as described previously (11).
Transplantation techniques
Mouse hearts were generally transplanted to a heterotopic abdominal location with appropriate microsurgical anastomoses according to our previously described technique (12). In one experiment, hearts were transplanted to a cervical site by a previously described vascular cuff technique (13).
The continuing status of transplanted hearts was determined by direct palpation at least twice weekly with the vigor of contractions of the transplants being recorded on a scale of 1–3+. Heart transplant recipients received no additional treatment after their original preparation except in one series of experiments. In these experiments, one group of C.B-17/SCID recipients of B10.BR heart transplants was treated on days -6, -3, and -1 before transplantation with anti-CD4 and anti-CD8 mAbs. This treatment was used to conform to our previous experiments with allogeneic heart transplants in which such suppression was necessary to permit long-term heart survival and the development of arteriopathy. Some isologous control transplant recipients also received this treatment. In a second group of animals, in addition to the CD4 and CD8 Abs, an Ab directed against macrophages and NK cells was given on day -1 and once weekly after transplantation for 8 wk. This Ab, anti-asialo GM-1, or ASGM-1, is a rabbit polyclonal reagent, obtained from WAKO (Richmond, VA). In some hands, this Ab has been reported to react quite selectively with NK cells (14).
Full-thickness skin grafts of flank or tail skin were placed upon recipient beds on the lateral thoracic walls preserving the panniculus carnosus muscle layer according to the standard technique described by Billingham and Medawar (15). Dressings were removed at 7 or 8 days and the status of the grafts was determined by inspection, and, in some instances, confirmed by histological examination. Long-term surviving grafts were readily distinguishable from surrounding skin as they maintained their original size and sustained vigorous hair growth.
Histological techniques
The transplanted hearts were removed from their recipients, frozen in OCT compound (Ames, Division of Miles Laboratories, Elkhart, IN), and stored at -20°C. Tissue blocks containing the hearts were placed upon the stage of a freezing microtome. Sections were cut at 4 µm, acetone or Formalin fixed, and routinely stained with H&E or elastic van Gieson. Frozen sections were also stained by immunoperoxidase using standard techniques. Immunoperoxidase-stained sections were then developed in a solution of 3-amino-9-ethyl carbazole (Aldrich, Milwaukee, WI) as described elsewhere (16), postfixed in 4% formaldehyde, counterstained with hematoxylin, and mounted in Gelvatol (Monsanto, Springfield, MA).
The severity and distribution of obstructive arterial lesions in the
coronary systems of transplanted hearts was generally evaluated on
frozen sections stained with the elastic tissue stain. In our
experience, the proximal coronaries are preferentially affected.
Therefore, efforts were made to include the proximal, major coronary
vessels about the base of the heart in the sections to be examined.
Proliferative changes in the coronary arteries were classified as stage
I, in which definite increased intimal cellularity and thickening were
apparent, but encroaching upon <50% of the vessel lumen, stage II,
including 
50% of the lumen, or stage III, occupying >50% of the
lumen. The number of coronary vessels available for evaluation in an
individual heart, and the number of these that were affected, was also
recorded. This system of evaluation has been described in detail
previously (9). In addition to focusing in this study upon
the larger coronary vessels, we also determined the anatomical
distribution of the lesions throughout the coronary systems. All
vessels sectioned through their lumena were scored for the presence or
absence of intimal thickening. Three levels were categorized:
"proximal" denotes coronary arteries at their origin;
"epicardial" refers to other coronary arteries in epicardial fat;
and "intramyocardial" are all arteries within the myocardium.
Controls for histological scoring were fully reactive B10.BR recipients
of B10.A hearts (n = 61) (9). Statistical
comparison of results, where needed, was performed using the Fisher
exact test.
Induction of neonatal tolerance or mixed chimerism
Neonatal tolerance was instituted by the i.v. injection into the recipient’s facial vein of 5–10 x 106 dissociated spleen cells from adult members of the F1 donor strain within 24 h of birth (5).
Mixed chimerism was instituted as described previously (8, 17) in age-matched (8- to 12-wk-old) mice after they had received 3 Gy total body irradiation and the i.v. injection, on the same day (day 0), of unseparated bone marrow harvested from their MHC-mismatched donors (8–12 wk old). Mice also received a range of T cell depletion doses of rat IgG2b anti-mouse CD4 mAb GK1.5 (0.25–1.0 mg) and anti-mouse CD8 mAb 2.43 (0.02–1.0 mg) on day -1. Cell clones for GK 1.5 and 2.43 were obtained from the American Type Culture Collection (Manassas, VA). Hamster anti-mouse CD40L mAb (MR1) was injected i.p. on day 0 (0.5 mg). The MR1 hybridoma was kindly provided to us by R. J. Noelle (Dartmouth Medical School, Lebanon, NH).
Mice in both the neonatally treated group, and in most which developed mixed chimerism, underwent donor strain skin grafting to substantiate their high degrees of nonreactivity to donor Ags. Those animals that fulfilled the criterion of graft survival in perfect condition for at least 100 days were admitted further to the experiment. This occurred in >80% of mice rendered tolerant at birth and almost 100% of mixed chimeras. Frequently grafts were allowed to survive considerably longer. To evaluate the possible influence of test skin grafting upon the immunological responses of chimeric recipients, some mice received heart transplants in the absence of prior skin grafting, it having been confirmed that donor-specific chimeric cells were present. In all cases, potential recipient mice that had undergone treatment to induce tolerance or mixed chimerism received no further treatment to alter immune reactivity. At the conclusion of each experiment in which donor cell chimerism was expected, in addition to examining each mouse for donor cells in the peripheral blood and/or spleen, specimens of the grafted skin (where applicable) and the native and transplanted hearts were removed for histological and immunopathological analysis.
Abs and flow cytometry
Flow cytometric analysis for the presence of multilineage chimerism was performed as previously described (7). In brief, forward angle and 90° light scatter properties were used to distinguish lymphocytes, monocytes, and granulocytes in peripheral white blood cells. Two-color flow cytometric analysis was used to distinguish donor and host cells of particular lineages, and the percentage of donor cells was calculated as previously described (18) by subtracting control staining from quadrants containing donor and host cells expressing a particular lineage marker, and by dividing the net percentage of donor cells by the total net percentage of donor plus host cells of that lineage. Dead cells were marked for exclusion using propidium iodide staining. Nonspecific FcR binding was blocked by anti-mouse FcR mAb 2.4G2 (18). FITC-conjugated mAbs included anti-CD4, anti-CD8, anti-B220 (all purchased from BD PharMingen, San Diego, CA), and anti-Mac1 (Caltag Laboratories, San Francisco, CA). Negative control mAb HOPC1-FITC, with no reactivity to mouse cells, was prepared in our laboratory. Biotinylated anti-H-2Dd mAb 34-2-12, anti-H-2Kk mAb 36-7-5 (BD PharMingen), and a control mAb HOPC1 were developed with PE-streptavidin.
Ab determinations
Complement-dependent cytotoxic Ab titers were determined by trypan blue dye exclusion using dissociated spleen cells as targets, as described previously (19). The presence of nonspecific Igs in the sera of recipients was assessed by a simple precipitin test (20). In this test, 20 µl of mouse serum was incubated in a microtiter plate with 5 µl of goat anti-mouse IgG1 or IgG2a serum for 1 h at 37°C. The plate was centrifuged for 1 min at 2000 rpm and allowed to stand at room temperature for 1 h. The precipitin reaction in each well was then recorded on a scale of 0–2+.
Immunopathology
Immunopathological analysis was performed in similar fashion to previous studies (9). In brief, cryostat sections were stained with Abs to mouse CD3, CD4, CD8, Mac1, and ASGM1 and to donor and recipient class I-determined Ags. The distribution and intensity of the infiltrate was noted for each of the markers. Sections were also stained with H&E and for elastin to determine evidence of vascular inflammation and fibrosis and of rejection activity within the myocardium.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Neonatally tolerant recipients
Table I
(group A) includes the skin
graft and heart transplant survival as well as the distribution and
prevalence of coronary artery lesions in 15 mice in the neonatally
induced tolerance group (B10.A to B10.BR). At the time of termination
of the experiments, donor cells bearing CD8 markers were detectable in
substantial numbers in the spleens of treated mice (in the 11–28%
range), although few such cells were detectable in the peripheral blood
(<1%; Table II, group A). Donor-derived
cells of other lineages were not present in substantial numbers.
Several skin grafts, each of which remained in perfect condition on
inspection, were also examined histologically (example below, Fig. 1). This revealed not only that each skin
graft remained completely free of infiltrating inflammatory cells, but
also that viable donor skin structures had persisted normally in the
graft. It will also be noted from Table I that the proximal coronary
vessels of 12 of 15 recipients examined showed distinct evidence of
arteriopathy, of varying degrees of severity, and that 13 of the 19
proximal coronaries examined showed advanced lesions (stages II and
III), all in the presence of surviving skin grafts from the donor
strain, as well as donor cell chimerism at a substantial level. In Fig. 1, we present these findings in detail for a single, representative
animal to exemplify their simultaneous presence in an individual
mouse.
|
|
|
Recipients with mixed chimerism
Similar results regarding evidence for a specific lack of
reactivity to donor Ags were obtained in the case of mice of the two
strain combinations made tolerant according to the mixed chimerism
protocol (B10.A to C57BL/6 and B10.A to B10.BR, see Table I
, groups B
and C). Thus, all of the mice that received skin grafts supported their
survival in perfect condition throughout the duration of the
experiment. Their levels of donor cell chimerism were generally higher
than in mice rendered neonatally tolerant (see Table II, groups B and
C), and a broader range of cell types was present.
Hearts were transplanted to a total of 19 fully tolerant recipients in
the two strain combinations in which mixed chimerism had been
established. In these mice, tolerance had been proved by the long-term
survival of a donor strain skin graft. The findings from these heart
transplants are summarized in Table I. Among recipients of fully
MHC-mismatched marrow (B10.A to BL/6), hearts transplanted to the
abdominal location in five of eight mice developed vasculopathy of
varying severity throughout their coronary systems. Among six hearts
transplanted to the neck, all six showed vasculopathy. In the class
I-only mismatched B10.A to B10.BR combination, two of five hearts
showed evidence of vasculopathy which was mild (stage I) in
severity and lesions in epicardial and intramyocardial portions of the
coronary system were absent.
Of four B10.A hearts transplanted to B10.BR mixed chimeric recipients,
which had previously been shown to contain donor-specific cells by flow
cytometric analysis, but had not undergone skin grafting, two
manifested typical arteriopathic lesions when their heart transplants
were examined at 30–91 days. Persistent donor cell chimerism was also
confirmed at this time (Tables I and II, group D). These results were
reassuring in regard to the possible perturbation of the system by the
presence of skin grafts in some chimeric recipients before the
placement of heart transplants.
Coronary lesions in the arteries of transplants to recipients
manifesting either neonatal tolerance or mixed chimerism revealed a
cellular infiltrate in the intima and adventia consisting of
CD3+, CD4+,
Mac1+, IA+, and
ASGM1+ cells, with few CD8+
cells (Fig. 2). More commonly, intimal
fibrosis with a scant cellular infiltrate and luminal narrowing was
evident, presumably representing a later stage of the inflammatory
process. There was little or no inflammation in the myocardium. The
cellular and fibrotic lesions were more frequent in the proximal graft
coronary arteries near their ostia and in the epicardial zone in all
groups, similar to the results found in previous studies using fully
reactive recipients (9) (see Fig. 4). The distal
intramyocardial arteries were affected in similar frequency in the
neonatally tolerant (14%), mixed chimeric (4%), and fully reactive
mice (13%).
|
|
Sixteen B10.BR hearts were transplanted to otherwise untreated
C.B-17/SCID recipients (shown in both Tables III and V). Of these, 11
were found to have definite proximal coronary vascular lesions when the
animals were sacrificed 29–64 days later. Similar lesions were found,
although less commonly, in hearts transplanted to
RAG1-/- and SCID/beige recipients, as detailed
in Table III. In most cases, the lesions
were fibrotic with a scant infiltrate. Those BR allografts in SCID and
RAG1-/- recipients whose lesions were more
cellular showed a coronary arterial intimal infiltrate including
Mac1+ and ASGM1+ cells.
Cells bearing markers for CD3+,
CD4+, and CD8+ were also
detected in the intima in some hearts (Fig. 3).
|
|
|
). To assess the possibility of some residual immune reactivity in SCID recipients, the sera of seven of these mice that had received heart transplants were tested for the presence of Ig. Two of these gave positive reactions. Similar tests of three SCID/beige recipients were all positive. This suggested, of course, that a degree of immunological "leakiness" had occurred in some SCID mice consistent with the presence of CD3+ cells. No relationship could be established between this finding and the presence or severity of coronary lesions. This shortcoming should not apply to RAG1-/- recipients, and no trace of Ig was found in several of these mice that were tested. To establish without question that RAG1-/- recipients were able to mount arteriopathic lesions in H-2-incompatible transplanted hearts, two mice in which such lesions had formed were confirmed as being bona fide RAG1-/- animals by PCRs showing both the presence of the neomycin marker and the absence of the RAG1 gene (tests kindly performed by J. Bracy and J. Iacomini, Transplation Biology Research Center, Massachusetts General Hospital, Boston, MA).
Isogeneic controls
Seven B10.A to B10.A isologous heart transplants and 16 B10.BR
isografts were performed to provide concurrent controls for the present
experiments. Six isologous transplants were also performed between
members of the unreactive strains (Table IV). All but two hearts were completely
free of arteriopathy at all levels of scoring, and those hearts showed
no more than slight evidence of vascular involvement.
|
The participation of "innate" reactivity, and particularly NK
cells, in the production of vascular lesions under these conditions was
considered to be an interesting possibility. Accordingly, nine B10.BR
hearts were transplanted into C.B-17/SCID/beige recipients. Five of
these transplanted hearts developed proximal coronary lesions (see
Table III).
In an additional experiment, 13 SCID recipients of B10.BR heart
transplants received anti-CD4 and anti-CD8 Abs before
transplantation and were also treated on day -1 and once a week
thereafter with anti-ASGM-1 Ab. This treatment significantly
reduced lesion formation in this group of recipients as compared with
animals in the same strain combination treated with anti-CD4 and
anti-CD8 Abs alone (p < 0.0083). In those
recipients treated with ASGM1 in which transplanted hearts still
manifested some coronary arteriopathy (5 of 13), a number of
ASGM1+ cells had escaped elimination, whereas
CD3+ cells were rare. Treatment only with
anti-CD4 and anti-CD8 did not suppress lesion formation
significantly compared with untreated animals, as detailed in Table V (p =
NS).
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Despite a demonstrated lack of specific immune reactivity on the part of their recipients, many transplanted hearts, in the presence of either type of tolerance, developed distinct proliferative lesions in their coronary arteries. For the purposes of this study, we have focused upon the status of the larger vessels as they are more extensively involved and more readily examined. When grades of involvement are important to an experiment, other approaches, including morphometric techniques, can clearly be helpful. Here, the mere presence of arteriopathy, in the circumstances we have examined, was taken as the main end point. All of these lesions involved the characteristic rich population of inflammatory cells in the linings of affected vessels along with proliferative and centripetal migratory responses by smooth muscle cells of the media. The coronary arteries of the allografts in both tolerance protocols showed both an advanced intimal fibrosis and a mixed mononuclear cell infiltrate, which included T cells, macrophages, and possibly NK cells, as judged by their reactivity with Abs to CD3, Mac1, and ASGM1. CD4+ cells and, to a lesser degree, CD8+ cells were present. These vascular lesions were similar in severity and distribution to those we have described in cardiac allografts in normal animals, although the cellular phase was less commonly observed in the tolerant animals and the frequency of involvement was somewhat diminished. In both the neonatal and mixed chimerism tolerance protocols, no evidence of acute rejection of the myocardium was observed as judged by a strong contractile pulse and the lack of histological injury and inflammation in the myocardium.
The finding of arteriopathy in tolerant recipients came as a surprise. Transplanted organs are usually permitted to survive in both experimental and clinical settings because of some form of continuing immunosuppressive treatment of their recipients. But this treatment may not fully suppress all immunological responses to the Ags in the donated tissue so that ongoing immune responsiveness under these conditions, and the presence of arteriopathy, is not unexpected. In our tolerant animals, arteriopathy was not anticipated because of ample evidence that chronic changes in transplanted organs are generally attributable to an ongoing specific immune response to them on the part of their recipients, as mentioned above. Evidence for this includes extensive localization of T lymphocytes and other inflammatory cells in these lesions and the dependence for their production on incompatibility between donors and recipients. Furthermore, vasculopathic lesions can be produced by the passive transfer of Abs specifically reactive with donor Ags (21) or by the transfer of appropriate immunologically competent cells (22).
In considering mechanisms for the appearance of vascular lesions in organs transplanted to tolerant recipients, several possibilities were entertained. First, the tolerant states in question, although they seemed to be quite profound by all our tests, might not have been absolute, and a small degree of continuing specific reactivity could have been spared by the treatments given. The evidence for robust and systemic tolerance described above makes this explanation seem improbable. Second, the tolerant states concerned that had been instituted by lympho-hemopoietic cells might not have extended to all of the foreign Ags presented by transplanted hearts. Thus, genetically segregating Ags in the transplanted hearts, perhaps expressed especially in certain vessels, might have continued to provoke an immune response. This could have occurred even though full tolerance of all of the Ags present in the lympho-hemopoietic cells of the donor inocula and also in the skin grafts placed later in the strain combinations selected had been achieved.
In any case, either of these two possibilities would depend upon some form of continuing immune responsiveness toward the donated tissues. Accordingly, we sought evidence regarding the fate of hearts transplanted to recipients rendered profoundly incapable of all immune responses by two different genetic alterations represented by C.B.17/SCID and RAG1-/- mice. Here, again, we have regularly found proliferative coronary vascular lesions in hearts transplanted to such recipients. Isotransplants between members of these deficient strains were generally free of coronary vascular lesions. Thus, in these experiments, coronary lesions were found principally in circumstances in which the donors and recipients concerned were incompatible with one another. Some evidence of leakiness as manifested by the presence of Igs was detected in 6 of 14 SCID and SCID/beige mice. Even in these mice immune reactivity must have been quite profoundly impaired, however, and no systematic relationship between the presence of detectable Igs and the presence of lesions in transplanted hearts was evident. Arteriopathy was observed in two animals in which no Igs could be found. In addition, in vivo depletion with anti-CD4 and CD8 Abs had no impact on the incidence of these lesions. Nevertheless, this made observations with RAG1-/- recipients all the more useful as no evidence for leakiness exists in this strain. The coronary artery lesions in both SCID and RAG1-/- mice were similar to those found in recipients in both tolerance protocols, even to the presence of scattered CD3+ cells. Cells reacting with Mac1 and ASGM-1 were particularly prominent in the cellular phase of arteriopathy. The origin of the cells (donor or recipient) was not determined, although the presumption is that the T cells in the RAG1-/- mouse must have been of donor origin; CD3+ cells have been described in SCID animals in small numbers (23, 24).
We do not take these findings to indicate that arteriopathy in transplanted organs generally occurs apart from specific immunological reactivity. Indeed, we continue to believe that there is convincing evidence that immunological responses to donor-specific histocompatibility Ags are an important instigating cause for these changes in most circumstances. What we consider has been raised by the above results is that conventional immunological reactivity to donor-specific Ags need not be the only cause for arteriopathy in transplanted organs between incompatible individuals. Other mechanisms could be responsible for arteriopathic lesions, either along with specific immunological responses or independent of them.
What could explain such inflammatory lesions in the absence of specific
T and B cell immunity? A third pathway of primitive or innate
responsiveness, such as that by the cytotoxic or cytokine release
properties of NK cells, could play a part. We selected
C.B-17/SCID/beige recipients for further tests because their NK cells
have been demonstrated to be impaired as regards certain functions
(25). The beige defect is the murine counterpart of the
human Chédiak-Higashi syndrome, which is a defect in lysosomal
granule exocytosis that impairs NK cell cytolytic activity. This defect
would not preclude the ability of such NK cells to produce IFN-
and
other cytokines upon activation. In previous studies to address the
question of whether or not donor and host NK cells are "tolerant"
of one another in mixed chimeras, it was found that recipient NK cells
from them are able to kill donor lymphoblast targets after stimulation
in vitro with IL-2 and donor stimulator cells, yet they maintain
tolerance of self-Ags (26). On the other hand, standard in
vivo assays for NK cell-mediated marrow rejection in established mixed
chimeras also suggest, however, that host NK cells are indeed tolerant
of the donor (Y. Zhao and M. Sykes, unpublished data). Nevertheless,
these cells may still be the sources of certain cytokines, including
IFN-.
Therefore, our findings in C.B-17/SCID/beige recipients should
not weigh decisively against the participation of NK cells particularly
in view of the known importance of IFN- in the production of
arteriopathy. We found some time ago that blocking the effects of
IFN- with a mAb specifically reactive to it would significantly
inhibit the development of arteriopathic lesions (27), and
others have since found that treatment of mice with IFN- alone can
produce them (28).
The reduction in severity of arteriopathy in SCID recipients following
treatment of heart transplant recipients with ASGM-1 Ab is consistent
with the possibility that NK cells are involved in the process. The
glycolipid ASGM-1 is expressed by NK cells, and the polyclonal
anti-ASGM-1 has been reported to deplete NK activity specifically
(14). ASGM-1 is also expressed, however, by dendritic
cells, macrophages, and cytotoxic T cells (29, 30).
Treatment with anti-ASGM-1 in vivo can suppress NK function without
affecting T cell immune reactions (31), although
inhibition of cytotoxic T cells has been reported (32).
Paradoxically, anti-ASGM-1 increases Ig levels in SCID mice
(24). The lack of precise specificity of this polyclonal
Ab for mouse NK cells must leave the question open for further
definition. It might seem unlikely that NK cell activity, presumably
via IFN-, could be mainly responsible for the inflammatory lesions
we have observed, but it should be borne in mind that the process of
arteriopathy is a chronic one that could well occur as a consequence of
quite low-grade reactivity over a considerable period of time.
In the context of possible NK cell activity in the present
circumstances, how can the difference in behavior of skin grafts vs
heart transplants be explained? Skin grafts have not been shown to be
susceptible to attack by NK cells, and this conclusion has also been
reached for transplanted organs (33). More recently,
however, evidence has been advanced supporting involvement of NK cells
in acute organ rejection (e.g., Ref. 34). It may also be
relevant that long-term surviving skin grafts can undergo gradual
replacement of endothelial cells in small vessels by recipient-derived
cells (35), whereas replacement of endothelial cells in
transplanted organs is usually limited (36), although some
reports suggest it to be present in varying degrees (37, 38). On the other hand, considerable evidence now supports the
view that the derivation of the majority of neointimal cells, bearing
-actin markers characteristic of smooth muscle cells, is from the
recipient, although their origin, in different situations, may be
either from bone marrow (39, 40) or from adjacent vessel
walls (41).
The involvement of various infectious agents in the generation of conventional atherosclerosis has been suspected for some time, although careful recent studies have led to the conclusion that their presence should not be considered a necessary component of the process (42). Nevertheless, one could suspect that some infectious agent might have found its way into the tissues of either some of our donors or recipients and be implicated along with foreign histocompatibility factors carried by the donor animals. To explore this possibility, a group of mice, in the combination B10.A donors and C57BL/6 recipients, was housed and treated throughout in a specific pathogen-free environment. These mice underwent the standard protocol for production of mixed chimerism. Five of six in this group of heart transplants developed coronary vascular lesions. This evidence argues against the involvement of ordinary pathogens in the process under study.
Several reports exist regarding the effects of various forms of specific unreactivity, or tolerance, on the formation of arteriopathic lesions in transplanted organs. In two, tolerant states were induced in respect to donor Ags determined by genes of the MHC between rats (43) or miniature swine (44). In the former case, tolerance followed infusion of donor bone marrow cells and a course of cyclosporine, whereas in the latter it was induced between swine differing at class I of the MHC following a short course of cyclosporine and a donor-specific kidney transplant. In both, the test organ was a heterotopic heart transplant. No evidence of arteriopathic lesions was found in either of these experiments. When immune responsiveness is impaired by interfering with the "second signal" of activation in, for example, CD40 ligand-deficient recipients, even though transplant survival is extended, arteriopathy still occurs (45). A combination of treatment with a mAb to CD40 ligand plus donor cells has, however, been reported to eliminate arteriopathic changes in transplanted mouse hearts (46).
Two reports are also available in which tolerance, induced to non-MHC-determined Ags, likewise appeared to be sufficient to prevent the formation of arteriopathic lesions in transplanted hearts. One made use of mice (AKR and C3H) differing in respect to Mls-1 and multiple minor histocompatibility Ags, but not for H-2-determined Ags (47) and the other of rats of the F344, ACI, and WF strains (48). More recently, further observations have come to light from one group that arteriopathic lesions can, indeed, be found in swine hearts transplanted to tolerant recipients under the conditions of their experiments (49).
These inconsistencies in results are not easy to account for, although variability in the vigor of innate responsiveness in different settings and variations between species, along with the time it may take for these chronic changes to occur under such widely different situations, could contribute. Different forms of nonreactivity also may have different outcomes, and it remains possible that certain forms of "peripheral tolerance" can, for example, lead to different results than we have found with the profound "central" tolerance we have been investigating. Indeed, of the two forms of tolerance tested here, there is a suggestion that arteriopathic changes may be somewhat less prevalent in recipients made tolerant through mixed chimerism than by neonatal exposure to donor-specific Ags. Furthermore, the mechanisms involved in producing nonreactivity, even of neonatal tolerance, have recently come into question as evidence for the importance of a preferential Th2 response rather than a central deletion mechanism has been advanced (50, 51).
In any case, we can assert that arteriopathic lesions were clearly present in major coronary arteries under the circumstances of our experiments. Although a full understanding of the mechanisms at work in these complex settings will require additional studies, the present results make it clear that incompatibilities, at least of MHC-determined Ags, are sufficient to provoke the formation of arteriopathy in the absence of any demonstrable conventional immune activity, either cellular or humoral. As mentioned above, a number of pathways toward the formation of arteriopathic lesions are known. The experiments reported here suggest that, at least in certain circumstances, additional possibilities, including innate responsiveness, should be added to the other mechanisms already known as possible causes for transplant arteriopathy.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Preliminary results of these experiments were presented in brief at a combined meeting of the American Society of Transplant Surgeons and the American Society for Transplantation, Chicago, IL, May 17, 2000.
3 Address correspondence and reprint requests to Dr. Paul S. Russell, Transplantation Unit, White 510, Department of Surgery, Massachusetts General Hospital, Boston, MA 02114. E-mail address: psrussell{at}partners.org
4 Abbreviation used in this paper: AS, asialo.
Received for publication February 23, 2001. Accepted for publication September 18, 2001.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
. Transplantation 57:1367.[Medline]
elicits arteriosclerosis in the absence of leukocytes. Nature 403:207.[Medline]
This article has been cited by other articles:
|
|
 |
|
  J. N. Beilke and R. G. Gill Frontiers in Nephrology: The Varied Faces of Natural Killer Cells in Transplantation Contributions to Both Allograft Immunity and Tolerance J. Am. Soc. Nephrol., August 1, 2007; 18(8): 2262 - 2267. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Forman, E.-S. Kang, C. Tian, J. Paez-Cortez, and J. Iacomini Induction of Alloreactive CD4 T Cell Tolerance in Molecular Chimeras: A Possible Role for Regulatory T Cells J. Immunol., March 15, 2006; 176(6): 3410 - 3416. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Uehara, C. M. Chase, W. H. Kitchens, H. S. Rose, R. B. Colvin, P. S. Russell, and J. C. Madsen NK Cells Can Trigger Allograft Vasculopathy: The Role of Hybrid Resistance in Solid Organ Allografts J. Immunol., September 1, 2005; 175(5): 3424 - 3430. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. I. Skaro, R. S. Liwski, J. Zhou, E. L. Vessie, T. D.G. Lee, and G. M. Hirsch CD8+ T cells mediate aortic allograft vasculopathy by direct killing and an interferon-{gamma}-dependent indirect pathway Cardiovasc Res, January 1, 2005; 65(1): 283 - 291. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Metzler, P. Gfeller, M. Bigaud, J. Li, G. Wieczorek, C. Heusser, P. Lake, and A. Katopodis Combinations of Anti-LFA-1, Everolimus, Anti-CD40 Ligand, and Allogeneic Bone Marrow Induce Central Transplantation Tolerance through Hemopoietic Chimerism, Including Protection from Chronic Heart Allograft Rejection J. Immunol., December 1, 2004; 173(11): 7025 - 7036. [Abstract] [Full Text] [PDF] |
