Development of a Combined Heart and Carotid Artery Transplant Model to Investigate the Impact of Acute Rejection on Cardiac Allograft Vasculopathy

EXPERIMENTAL TRANSPLANTATION

Development of a Combined Heart and Carotid Artery Transplant Model to Investigate the Impact of Acute Rejection on Cardiac Allograft Vasculopathy Behzad Soleimani, MD, MRCP, FRCS(C-Th),a Fumin Fu, MD,b Philip Lake, PhD,b and Victor C. Shi, MDb Background: Cardiac allograft vasculopathy (CAV) is the leading cause of late allograft loss after heart transplantation. Although clinical studies are suggestive of an association between episodes of acute rejection and subsequent emergence of CAV, direct experimental evidence in support of a causal relationship is lacking. Methods: We developed a new murine model of CAV in which a carotid artery and a heart graft are simultaneously transplanted into a single recipient. Transplants were performed across full or partial major histocompatibility complex (MHC) mismatched strain combinations. The heart grafts were either syngeneic with the carotid graft or from a third-party strain. Carotid arteries were harvested after 30 days and evaluated by morphometry and immunohistochemistry. Results: In the fully mismatched combination, all heart grafts were rejected within 7 days, as determined by loss of pulsation. At 30 days, carotid allografts in the combined transplant group had significantly more intimal hyperplasia compared with isolated carotid allografts. The neointima consisted of abundant smooth muscle cells and leukocytes. Intimal hyperplasia was also significantly enhanced by acute rejection of the third-party donor heart. In the partial MHC mismatched combination, the heart graft survived indefinitely, and this was associated with diminished intimal hyperplasia in the cotransplanted carotid artery compared with the isolated carotid allograft. Conclusion: We present direct experimental evidence that CAV is promoted by acute parenchymal rejection of the heart. This interaction between acute rejection and CAV is mediated by both allospecific and non-allospecific processes. Effective therapeutic strategy against CAV should therefore target non-allospecific mediators as well as prevent episodes of acute rejection. J Heart Lung Transplant 2008;27:450 – 6. Copyright © 2008 by the International Society for Heart and Lung Transplantation.

Cardiac allograft vasculopathy (CAV) is the leading cause of heart transplant failure after the first postoperative year.1,2 The key pathologic process in CAV is the rapid development of intimal hyperplasia (IH). The development of IH is characterized by the accumulation of smooth muscle cells (SMCs) and extracellular matrix in a subendothelial location. This occurs together with infiltration of monocytes, T cells, fibroblasts, and dendritic cells.3,4 A number of experimental models have been de-

From the aDepartment of Cardiac Surgery, National Heart and Lung Institute, Imperial College London, London, United Kingdom, and b Novartis Pharmaceuticals Corporation, East Hanover, New Jersey. Submitted July 26, 2007; revised December 23, 2007; accepted January 12, 2008. This work was funded by grants from the British Heart Foundation and Novartis Pharmaceutical Corporation. Reprint requests: Victor C. Shi, Novartis Pharmaceuticals Corporation, One Health Plaza, East Hanover, NJ 07936-1080. Telephone: 862-778-8300. Fax: 973-781-8265. E-mail: [email protected] Copyright © 2008 by the International Society for Heart and Lung Transplantation. 1053-2498/08/$–see front matter. doi:10.1016/ j.healun.2008.01.015

450

veloped to determine the cellular and molecular mediators of this complex process. Experimental heart transplants have long been recognized to develop CAV provided early acute rejection (AR) is prevented, generally by the use of immunosuppressants. Introduction of immunosuppressants can, however, increase the complexity of the models given that many of these agents have themselves been implicated in the pathogenesis of CAV.5–7 In addition, immunosuppressants used experimentally generally have no clinical application and in some models result in permanent graft survival after a brief period of peri-operative treatment.8 This is clearly not the case in clinical transplantation, and this discrepancy may have important implications when extrapolating results from these models to clinical CAV. An alternative strategy is the use of partial major histocompatibility complex (MHC) or minor histocompatibility mismatched strain combinations. Because in clinical transplantation, donors and recipients are often mismatched in multiple major and minor histocompatibility loci, data generated by heart transplant models

The Journal of Heart and Lung Transplantation Volume 27, Number 4

Soleimani et al.

451

Table 1. Donor and Recipient Strains in Experimental and Control Groups Group I II III Control I Control II

No. 6 6 6 6 6

Recipient strain C3H(H-2k) B10.BR(H-2k) C3H(H-2k) C3H(H-2k) B10.BR(H-2k)

Carotid artery donor strain

using partial mismatch may only be applicable to a subset of patients with similar histoincompatibilities. One further characteristic of these models is that in common with clinical CAV, the coronary artery lesions in experimental cardiac allografts are generally heterogeneous in location, distribution, and intensity,9 making quantitative evaluation of IH severity in these models complex.10 In the most common variant of experimental heart transplantation, in which the heart is grafted heterotopically, the left ventricle is completely offloaded and undergoes atrophy. This results in the formation of left ventricular thrombus, which can in turn serve as a source of cytokines such as transforming growth factor-␤.11 The impact of left ventricular atrophy and thrombus formation on development of IH in these models is unknown. Arterial allograft models were developed as an alternative to whole organ grafts to allow mechanistic study of IH without the complications of an immune response to cardiac parenchymal tissue. In contrast with organ allograft models, the lesions seen in these models are concentric, uniform, and reproducible. In addition, most vessel transplant models require no immunosuppression because there is no acute destructive parenchymal rejection that would otherwise precede the emergence of IH. Absence of immunosuppression is, however, regarded by some authorities as a potential weakness of these models.12 To benefit from the advantages and avoid the shortcomings associated with the arterial and organ models, we developed a new model by cotransplanting a carotid artery and heart graft into a single recipient in the absence of immunosuppression. Advanced IH was reproducibly formed in the carotid artery graft within 30 days of transplantation. This model was then used to address the impact of acute rejection of the heart on the development of IH in the cotransplanted carotid artery. MATERIALS AND METHODS Animals and Experimental Groups Inbred mice (28 to 32 g, 9-week-old males) were chosen for incompatibility in the H-2 region (Jackson Laboratory, Bar Harbor, ME). Three groups of combined transplants and 2 groups of isolated transplants were performed (n ⫽ 6 per group). In Group I, C57BL/10 (H-2b) mice were used as donors of heart and carotid

b

C57BL/10(H-2 ) B10.A(2R)(H-2h2) C57BL/10(H-2b) C57BL/10(H-2b) B10.A(2R)(H-2h2)

Heart donor strain C57BL/10(H-2b) B10.A(2R)(H-2h2) BALB/c(H-2d) ●●● ●●●

artery grafts and C3H (H-2k) mice as recipients (full MHC mismatch). In Group II, B10.A(2R)(H-2h2) mice were used as donors of heart and carotid artery grafts and B10.BR(H-2k) mice as recipients (isolated class I MHC mismatch). In Group III, C3H(H-2k) mice were recipients of a carotid artery graft from C57BL/10(H-2b) mice and a heart graft from BALB/c(H-2d) mice (thirdparty heart). Isolated carotid grafts were also performed for Groups I and II as controls. The experimental and control groups are summarized in Table 1. All animals received humane care in compliance with the Principles of Laboratory Animal Care formulated by the National Society for Medical Research and the Guide for the Care and Use of Laboratory Animals prepared by the Institute of Laboratory Animal Resources and published by the National Institutes of Health (NIH Publication No. 86-23, revised 1996). Transplantation The cotransplant model used in this study was generated by transplanting carotid artery and heart grafts simultaneously into 1 recipient. The 2 procedures were performed under the same anesthetic, but in sequence with the heart, followed by the carotid artery. All procedures were performed on anesthetized mice under a dissecting microscope (Leica, Solms, Germany). Mouse carotid artery transplant model. This model has been described by us previously.13 In the donor mouse, a mid-line incision was made extending from the mandible to the suprapubis; then, 2 ml of heparinized saline (50 U/ml) was injected through the inferior vena cava, and the aorta was divided. The carotid arteries that lie laterally to the trachea were removed and stored in saline at 4°C. A mid-line incision was then made from the mandible to the suprasternal notch of the recipient. The left internal carotid artery was dissected out, and proximal and distal microvascular clips (FE693K, Aesculap, Center Valley, PA) were applied. Two longitudinal arteriotomies (0.5 to 0.6 mm, 1 cm apart) were made on the carotid artery using the tip of a 30-gauge needle. The graft was then transplanted as a loop by constructing 2 end-to-side anastomoses using 11-0 continuous nylon suture. Prominent pulsations were visible in both the transplanted loop and the native vessel after

452

Soleimani et al.

clamps were released. The grafts were subjected to 30 to 40 minutes of ischemia. Mouse heterotopic heart transplantat model. The model used here was a modification of the technique described by Corry et al.14 Briefly, the donor heart was harvested and stored in cold saline solution until transplantation. The donor’s ascending aorta and pulmonary artery were anastomosed to the recipient abdominal aorta and inferior vena cava, respectively. The anastomoses were constructed using 10-0 continuous nylon suture. Transplanted hearts resumed regular sinus rhythm shortly after reperfusion. The total ischemia time did not exceed 35 minutes. Functional status of the heart allograft was assessed by daily abdominal palpation. Although the diagnosis of cardiac allograft rejection by this method can be inaccurate, given that the precise timing of allograft loss was not our primary objective, we confined our assessment to palpation alone rather than performing additional histology. Immunosuppression. No immunosuppression was given either before or after transplantation. Preparation of Histologic Specimens The carotid grafts were harvested 30 days after transplantation by cutting the transplanted loop from the native vessel at the suture lines. The proximal half of the loop was fixed overnight in 10% formalin, and the other half was frozen in powdered dry ice and embedded in medium (Optimal Cutting Temperature Compound, Miles Inc, Madison, WI) for immunohistochemistry. Histologic sectioning was begun at the center of the graft to avoid the effects of the suture line, producing 4-␮m-thick sections. Histomorphometry Sections were viewed under a Nikon Lapshot-2 microscope (Tokyo, Japan) equipped with a Sony 3CCD camera (Tokyo, Japan) and a television monitor. The neointimal area was measured by computerized planimetry on sections treated with the Verhoeff’s stain, as described previously.13 The intima was defined as the region between the lumen and the internal elastic lamina. Area measurements were made on 3 sections treated with Verhoeff’s elastin staining obtained approximately 150, 300, and 450 ␮m from the center of the graft using the Image-Pro software (Universal Imaging Corp, Sunnyvale, CA). Three individual measurements per graft were averaged such that a final neointimal area was assigned to each animal. Cells were counted in hematoxylin-stained paraffin sections using the Image-Pro software and were expressed as total number of nuclei within neointima and nuclear density. Three individual measurements for

The Journal of Heart and Lung Transplantation April 2008

each graft were averaged such that a final number of cells were assigned to each graft. Results are given as the mean intimal area or cell number per group ⫾ standard error (SE). The significance of the differences between the groups was calculated using the 1-tailed Student’s t-test, with Bonferroni correction for multiple comparisons. A value of p ⬍ 0.05 was considered significant. Immunohistochemistry Sections cut about 500 ␮m from the center of the graft were selected from each carotid artery graft and were stained with ␣-actin and CD45 monoclonal antibodies, as described previously.15 A rat monoclonal antibody against CD45 (PharMingen, San Diego, CA) and mouse monoclonal anti–␣-SMC actin (Sigma, St Louis, MO) were used. The target cells were immunolocalized using an indirect immunoenzymatic Vectastain ABC method (Vector Labs, Burlingame, CA). Briefly, tissue sections were deparaffinized, non-specific binding was blocked by avidin/biotin blocking kit (Vector Labs), followed by preincubation in 10% normal goat serum. Primary antibodies were applied. A biotinylated goat anti-rat or anti-mouse secondary antibody was applied. CD45 expression was visualized with a 3,3’ diaminobenzidine tetrahydrochloride (DAB) chromogen substrate, and ␣-actin–positive SMCs were identified using a Vector Red substrate kit. Sections for CD45 staining were counterstained in 1% methyl green or in Verhoeff’s elastin stain for ␣-actin–positive SMC. RESULTS Heart and Carotid Artery Cotransplant Model In this model, a heterotopic heart graft and a paratopic carotid artery graft were performed in the same recipient under a single anesthetic. A surgical success rate of more than 90% was achieved with graft ischemia time of 30 to 40 minutes. The cardiac allografts rejected acutely in all fully mismatched strain combinations (C57BL/10 to C3H and BALB/c to C3H) by post-operative Day 7, as expected. The heart grafts in the B10.A(2R) to B10.BR strain combination did not acutely reject and were beating normally at the end of the 30-day observation period. The outcome of heart grafts in our study concurs with previously published data.8,16 The cotransplanted carotid allografts, harvested after 30 days, had uniform concentric IH (Figure 1B). This neointima contained abundant SMCs typical of mature lesions as defined by Tanaka et al,10 as well as leukocytes (Figure 1C, D). In contrast, isografts showed no IH with intima consisting of single layer of endothelial cells (Figure 1A).

The Journal of Heart and Lung Transplantation Volume 27, Number 4

Soleimani et al.

453

simultaneously transplanted (Figures 2 and 3A). The neointima in the cotransplanted carotid allografts had increased cellularity (Figure 3B), indicating that the accelerated neointima formation was the result of enhanced cellular infiltration. This is in contrast with the situation in isolated carotid allografts, where neointimal cellular density is observed to decline when the observation period was extended to beyond 30 days.13 It must, however, be noted that increased cellularity does not necessarily indicate enhanced cellular infiltration. Increased cellularity could be from decreased apoptosis or increased proliferation after infiltration is complete. Loss of cellularity after 30 days could be from deposition of collagen as occurs in human CAV. Impact of Acute Rejection in the Third-Party Heart Allograft on Intimal Hyperplasia Figure 1. Intimal hyperplasia in carotid artery cotransplant model. Photomicrographs show cross sections of 30-day mouse carotid artery allografts transplanted simultaneously with a heart from the same donor. C57BL/10 (H-2b) mice were used as donors and C3H (H-2k) as recipients. Elastic lamina is stained black (Verhoeff stain) in (A) isograft and (B) allograft. Allografts are shown with (C) immunohistochemical stain for smooth muscle cells (red) and (D) leukocytes (brown). Original magnification ⫻200.

The Impact of Acute Rejection on Intimal Hyperplasia In isolated carotid artery allografts, a moderate sized neointima was observed 30 days after transplantation in the C57BL/10 to C3H strain combination. In marked contrast, advanced and near-occlusive neointima was formed in almost all carotid artery allografts with which a heart graft from same donor strain was

When a heart graft from a BALB/c donor was cotransplanted with a carotid artery graft from a C57BL/10 donor into a C3H recipient, acute rejection of the third-party heart graft exacerbated IH in the carotid allograft. The severity of transplant IH was, however, less than that seen when the heart and carotid artery were from the same donor (Figures 2 and 4). The Impact of Cardiac Allograft Survival on Intimal Hyperplasia Isolated carotid artery allografts in the B10.A(2R) to B10.BR strain combination developed marked IH. However, cotransplantation of a heart and carotid artery in this strain combination, in which the heart graft survives indefinitely, resulted in diminution of IH in the carotid artery (Figures 2 and 4).

Figure 2. Intimal hyperplasia in isolated and cotransplanted carotid artery allografts. Photomicrographs show cross-sections of (top row) isolated carotid grafts and (bottom row) cotransplanted 30-day mouse carotid artery allografts in the (left panels) B10A(2R) to B10.BR, (middle pannels) C57BL/10 to C3H, and (right panels) C57BL/10 to C3H with heart graft from BALB/c mice, strain combinations. Original magnification ⫻200. Elastic lamina is stained black (Verhoeff stain) delineating the neointima.

454

Soleimani et al.

Figure 3. Bar charts show significantly greater (A) neointimal area and (B) neointimal total cell count in cotransplanted vs isolated carotid allografts (p ⬍ 0.001). The strain combination was C57BL/10 to C3H. The model was paratopic carotid artery transplantation, in the presence (combined) or absence (isolated) of a cotransplanted heart from same donor strain. The intimal areas are presented as mean in ␮m2 ⫾ SE. The cell count is expressed as the total number of nuclei in the neointima ⫾ SE.

DISCUSSION Although the precise mechanism of CAV remains undefined, AR has been suspected as a possible culprit on the basis of a number of clinical studies that have suggested that the 2 processes may be interrelated and that AR may be a trigger for subsequent development of CAV. Clinical studies that have addressed this question have yielded conflicting results, however, because of their limitations such as retrospective design, small study size, varying immunosuppression regimens used, and the method used to diagnose CAV. In most cases the diagnosis of CAV was based on angiography, a diagnostic method known to underestimate the presence and severity of CAV.17 Intravascular ultrasound performed serially has allowed careful evaluation of the severity and progression of CAV. A 3-year follow-up study of 47 cardiac allograft

The Journal of Heart and Lung Transplantation April 2008

recipients that used this diagnostic modality showed that the presence of acute cellular rejection was the only variable that correlated with the rate of progression of CAV.18 In a more recent 1-year angiographic study of 489 cardiac transplant recipients, acute moderate to severe cellular rejection had a cumulative impact on CAV onset, whereas mild, untreated rejection was not associated with CAV.19 An association between AR and CAV is also supported by a recent clinical study showing that the proliferation signal inhibitor everolimus was effective in reducing acute rejection as well as preventing CAV.20 Although many animal models have been developed to study the underlying mechanisms of IH in CAV, the question of association between acute cellular rejection and IH has remained largely unanswered because of the limitations imposed by these models. To address this issue, the combined transplant model was devised. The advantage of this model is the ease of assessing the severity of IH in the carotid allograft, in the context of an immune response mounted against the whole heart, in the absence of immunosuppression. In this model, we showed that the development of IH in the carotid allograft was significantly enhanced when the carotid artery was cotransplanted with a heart graft, which was acutely rejected shortly after transplantation. Although the mechanisms for this observation were not investigated and remain a matter of speculation, it is likely that the increased production of cytokines and growth factors from activated T lymphocytes during acute rejection21 may enhance the proliferation and migration of vascular SMCs, thus promoting the development of IH. It is conceivable that this interaction between AR and IH at cellular level has a significant non-cognate component, as demonstrated by the thirdparty heart experiment. An alternative explanation for the enhanced IH in the third-party cotransplant model is that the 2 donor strains may share minor histocompatibility antigens. This immunologic cross-reactivity may promote IH in the cotransplanted carotid artery allograft. Future experiments using different strain combinations in this model may help to clarify this issue. In striking contrast, the development of IH in carotid allograft was attenuated when the cotransplanted heart graft from B10.A(2R) was indefinitely accepted by the B10.BR recipient (class I mismatch). The underlying mechanism involved was not fully elucidated in the current study and will require further investigation. It is, however, possible to speculate a role for regulatory T cells in this process. Thus, dendritic cells arising from the donor heart present MHC class I allopeptides through their shared MHC class II molecules to selfrestricted regulatory T cells, promoting a state of tolerance. Because regulatory T cells have indirect allospeci-

The Journal of Heart and Lung Transplantation Volume 27, Number 4

Soleimani et al.

455

Figure 4. Bar chart shows the impact of the co-transplanted heart graft on intimal hyperplasia (IH) in the carotid artery allograft. In the C57BL/10 to C3H strain combination, simultaneous transplantation of a heart and carotid artery results in a marked enhancement of IH. In the B10A(2R) to B10.BR combination, where the strains differ in a single major histocompatibility complex class I locus alone, simultaneous heart transplantation results in mitigation of IH. Simultaneous transplantation of a BALB/c heart and a C57BL/10 carotid artery into a C3H recipient resulted in augmentation of transplant IH in the carotid artery. The areas are presented as mean in ␮m2 ⫾ SE.

ficity,22 they may protect carotid allografts from IH, which in this strain combination is likely to be indirect pathway driven. The isolated carotid allograft is, presumably, not as rich in dendritic cells as the whole heart and therefore does not have the same tolerogenic potential. In this setting, the indirect pathway prevails and results in a more pronounced IH. This situation may be analogous with the clinical situation described by Lagaaij et al,23 who showed that the survival of human cardiac and renal allografts could be improved by prior transfusion of whole blood from a donor sharing at least 1 human leukocyte antigen HLA-DR antigen with the recipient, in contrast with the deleterious effect of MHC-mismatched blood. This observation was explained by postulating the presentation of MHC class I allopeptides of the graft by blood-donor dendritic cells through the shared MHC class II molecules that are recognized by self-restricted regulatory T cells. In light of the increasing clinical evidence supporting a direct correlation between episodes of acute cellular rejection and CAV,2,18,19,24 a model that allows mechanistic study of this interaction without the use of immunosuppression may be of tremendous value. Future studies using this new model, which integrates the merits of arterial and organ transplant models, should provide further insight into the relative contribution of factors other than AR to the development of CAV. A better understanding of the immunologic mechanism involved in CAV will be important for the design of effective therapeutic strategies to ameliorate this process and to ensure long-term graft survival.

REFERENCES 1. Billingham ME. Graft coronary disease: the lesions and the patients. Transplant Proc 1987;19(4 suppl 5):19 –25. 2. Costanzo MR, Naftel DC, Pritzker MR, et al. Heart transplant coronary artery disease detected by coronary angiography: a multiinstitutional study of preoperative donor and recipient risk factors. Cardiac Transplant Research Databasem. J Heart Lung Transplant 1998;17:744 –53. 3. Newby AC, Zaltsman AB. Molecular mechanisms in intimal hyperplasia. J Pathol 2000;190:300 –9. 4. Ross R. Atherosclerosis--an inflammatory disease. N Engl J Med 1999;340:115–26. 5. Arai S, Teramoto S, Senoo Y. The impact of FK506 on graft coronary disease of rat cardiac allograft--a comparison with cyclosporine. J Heart Lung Transplant 1992;11:757– 62. 6. Mennander A, Tiisala S, Paavonen T, Halttunen J, Hayry P. Chronic rejection of rat aortic allograft. IIAdministration of cyclosporin induces accelerated allograft arteriosclerosis. Transpl Int 1991;4:173–9. 7. Mehra MR, Ventura HO, Chambers R, et al. Predictive model to assess risk for cardiac allograft vasculopathy: an intravascular ultrasound study. J Am Coll Cardiol 1995;26:1537– 44. 8. Russell PS, Chase CM, Winn HJ, Colvin RB. Coronary atherosclerosis in transplanted mouse hearts. ITime course and immunogenetic and immunopathological considerations. Am J Pathol 1994; 144:260 –74. 9. Armstrong AT, Strauch AR, Starling RC, Sedmak DD, Orosz CG. Morphometric analysis of neointimal formation in murine cardiac allografts. Transplantation 1997;63:941–7. 10. Tanaka M, Fedoseyeva EV, Robbins RC. Graft coronary artery disease in murine cardiac allografts: proposal to meet the need for standardized assessment. J Heart Lung Transplant 2005;24:316 –22. 11. George JF, Pinderski LJ, Litovsky S, Kirklin JK. Of mice and men: mouse models and the molecular mechanisms of post-transplant coronary artery disease. J Heart Lung Transplant 2005;24:2003–14. 12. Orosz CG. Considerations regarding the contributions of B cells to chronic allograft rejection in experimental animal models. J Heart Lung Transplant 2000;19:634 –7.

456

Soleimani et al.

13. Shi C, Russell ME, Bianchi C, Newell JB, Haber E. Murine model of accelerated transplant arteriosclerosis. Circ Res 1994;75:199 –207. 14. Corry RJ, Winn HJ, Russell PS. Primarily vascularized allografts of hearts in mice. The role of H-2D, H-2K, and non-H-2 antigens in rejection. Transplantation 1973;16:343–50. 15. Shi C, Lee WS, He Q, et al. Immunologic basis of transplantassociated arteriosclerosis. Proc Natl Acad Sci U S A 1996;93: 4051– 6. 16. Ardehali A, Billingsley A, Laks H, Drinkwater DC Jr, Sorensen TJ, Drake TA. Experimental cardiac allograft vasculopathy in mice. J Heart Lung Transplant 1993;12:730 –5. 17. Kobashigawa JA, Miller L, Yeung A, et al. Does acute rejection correlate with the development of transplant coronary artery disease? A multicenter study using intravascular ultrasound. Sandoz/CVIS Investigators. J Heart Lung Transplant 1995;14: S221– 6. 18. Jimenez J, Kapadia SR, Yamani MH, et al. Cellular rejection and rate of progression of transplant vasculopathy: a 3-year serial intravascular ultrasound study. J Heart Lung Transplant 2001;20: 393– 8.

The Journal of Heart and Lung Transplantation April 2008

19. Stoica SC, Cafferty F, Pauriah M, et al. The cumulative effect of acute rejection on development of cardiac allograft vasculopathy. J Heart Lung Transplant 2006;25:420 –5. 20. Valantine H. Prevention of cardiac allograft vasculopathy with Certican (everolimus): the Stanford University experience within the Certican Phase III clinical trial. J Heart Lung Transplant 2005;24(4 suppl):S191–5; discussion S210 –1. 21. Krensky AM, Weiss A, Crabtree G, Davis MM, Parham P. T-lymphocyte-antigen interactions in transplant rejection. N Engl J Med 1990;322:510 –7. 22. Yin D, Fathman CG. CD4-positive suppressor cells block allotransplant rejection. J Immunol 1995;154:6339 – 45. 23. Lagaaij EL, Hennemann IP, Ruigrok M, et al. Effect of one-HLADR-antigen-matched and completely HLA-DR-mismatched blood transfusions on survival of heart and kidney allografts. N Engl J Med 1989;321:701–5. 24. Brunner-La Rocca HP, Schneider J, Kunzli A, Turina M, Kiowski W. Cardiac allograft rejection late after transplantation is a risk factor for graft coronary artery disease. Transplantation 1998;65: 538 – 43.