Coronary artery immunogenecity: a comparison between explanted recipient or donor hearts and transplanted hearts

Transplant Immunology 1993; 1:294-301

Coronary artery immunogenicity: a comparison between explanted recipient or donor hearts and transplanted hearts Patricia M Taylor, Marlene L Rose and Magdi H Yacoub Department of Transplant Immunology, National Heart and Lung Institute at Harefield Hospital

Abstract: Transplant associated accelerated coronary sclerosis (TX-ACS) is the most serious complication following heart transplantation. In order to elucidate the involvement of possible immune mechanisms, we have used immunocytochemistry to characterize the antigens present on the endothelium and the cells present within coronary arteries taken at time of surgery from patients who required retransplantation (n = 4) or at post mortem (n = 10) from transplant patients who died. Coronary arteries from unused donor hearts (n = 4) or heart transplant recipients whose original disease did not involve the coronaries (n = 8) were used as controls. Endothelium from all control coronary arteries strongly expressed MHC class I, DR antigen, PECAM, ICAM-1 and E-selectin. Expression of VCAM-1, DP and particularly DQ antigen was more patchy. There was no significant difference in the expression of MHC or adhesion molecules between control arteries and those from transplant patients. More cells were present in the intima and media of coronary arteries taken from transplanted hearts, particularly those with intimitis/TX-ACS, than were found in control coronaries. T cells were present in the intima of most coronaries, but the media of only the transplanted coronaries. Cells expressing prolyl 4-hydroxylase were more evident on coronaries from transplanted hearts. In conclusion, coronary arteries from hearts used for transplants appear to be highly immunogenic. This suggests the endothelium would need little, if any, stimuli to intitiate as well as be a target for an immune response after transplantation.

Introduction Transplant associated coronary artery disease (TX-ACS) is the most serious complication following cardiac transplantation.~affecting 6% of our patients at one year and progressing to 17% after three years. Other centres have reported a higher incidence.2 Although there are several possible risk factors (number of acute rejection episodes, type of immunosuppression, viral infection, serum lipid concentrations), there is little agreement about their relative importance.3 The integrity of the endothelium is a crucial factor in maintaining normal vessel function. It is likely that endothelial damage is the earliest event that initiates all forms of arteriosclerosis.4 Recently, anti-endothelial antibodies have been described in patients with TX-ACS, 5 but whether these antibodies are the primary cause or a marker of endothelial danmge has not been ascertained. Here we have examined coronary arteries for expression of Address for correspondence: Patricia M Taylor, Department of Transplant Immunology, Heart Science Centre, Harefield Hospital, Hayfield, Middlesex UB9 6JH, UK. ~'~ Edward Arnold 1993

two groups of molecules. Both would be expected to contribute to the immunogenicity of a vessel, if transplanted into an allogeneic recipient. Molecules of the major histocompatibility complex (MHC), i.e. HLA-DR antigens, will initiate an immune response by activating recipient CD4 T cells and HLA-ABC antigens are a potential target for recipient CD8 T cells. Various adhesion molecules have been described which are either present constitutively on endothelial cells (PECAM, ICAM-1) or can be induced by cytokines (ICAM1, VCAM-1, E-selectin). These adhesion molecules allow adhesion of leucocytes to endothelial cells, which is presumably an early and essential event preceding extravasatiou.6,7 We have previously reported heterogeneous expression of MHC and adhesion molecules between different vessels of the human cardiovascular system,s This study included five coronary arteries from unused hearts that had been offered as donors for cardiac transplantation. We reported the surprising finding that donor coronary artery endothelial cells strongly expressed MHC class II antigens, raising the question whether these had been cytokine induced as a result of stress or trauma. Here we have extended these investigations

Coronary artery immunogenicity 295

to include coronaries from further unused donor hearts and compared them to transplant recipient explanted hearts whose original disease excluded coronary involvement. In order to investigate the effect of transplantation on antigen expression, we have examined coronary arteries from the transplanted heart of transplant recipients who had died or required retransplantation. We have addressed the question of the immunogenicity of coronary arteries prior to transplantation and the effect of transplantation on expression of MHC and adhesion molecules. In addition, we have characterized the cells present within the intima and media of control and transplanted coronary arteries.

Objectives The objectives of this work were: 1) To determine the expression of MHC antigens and vascular adhesion molecules on endothelium of control, i.e. nontransplanted coronary arteries; 2) To determine whether there is a difference in expression of MHC and adhesion molecules after transplantation; 3) To determine the phenotype of cells present within the intima and media of control and transplanted coronary arteries.

Materials and methods Twelve control coronary arteries (10 epicardial, two myocardial) were studied. Four were from donor hearts that were not used for transplantation. The reasons given were incompetent bicuspid aortic valve, hypotension, inotrope overload and steroid-induced oedema. Eight were from heart transplant recipients whose original disease did not involve the coronaries. These consisted of dilated cardiomyopathy (six) and congenital heart disease (two). The mean age of the

control group was 37 (range 17-56 years). Fourteen transplanted coronary arteries (13 epicardial, one myocardial) were studied. Four were from transplant recipients who required retransplantation and 10 were from transplant recipients at post mortem. The mean age of the transplant group was 33 (range four to 59 years). The time interval between initial transplant and retransplantation or death ranged from 12 days to five years six months. Eight patients had heart/lung transplants and six had heart transplants. The reason for retransplantation was obliterative bronchiolitis on three occasions, two of which also had TX-ACS or on one occasion due to surgical problems. Cause of death was infection (7), multiorgan failure (2) and TX-ACS (1). Only five of the 14 transplanted coronaries showed intimal cellular proliferation typical of TX-ACS. Two of these also showed evidence of intimitis as did one other specimen. Specimens were snap frozen and stored in liquid nitrogen until required. Frozen sections (6 p,m) were air-dried, fixed in acetone and stained with mouse monoclonal antibodies (Table 1). The cells binding antibody were visualized using immunoperoxidase. The primary monoclonal antibody was followed by biotinylated antimouse immunoglobulins (Dakopatts) and then by avidin--biotin-horseradish peroxidase complex (Dakopatts). Sites of peroxidase fixation were identified by incubation in phosphate buffered saline containing diaminobenzidene tetrahydrochloride (0.25 mg/ml) and hydrogen peroxide (0.01%). The sections were counterstained with Harris' haematoxylin. When the primary antibody was omitted there was an absence of staining. Specimens were given a value of l(min.) to 3(max.) for intensity and l(min.) to 4(max.) for proportion of endothelium stained. These values were totalled to give a score for overall expression of antigen by the endothelium lining the coronary arteries (Tables 2 and 4). Calls present on the luminal surface, within the intima or within the media of the coronary arteries were given a value of l(min.) to 10(max.) (i.e. all cells positive). The average

Table 1 Primary monoclonal antibodies used in immunocytochemistry MAb

Specificity

Reference

Mon0002 Mon0001 M616

Endothelium (EN4) Some endothelium (Pal-E) von Willebrand Factor (vWF)

Bradsure Biologieals Bradsure Biologicals Dakopatts

w6/32 L243 B7.21 Leu I(P

Non-polymorphic, MHC class I DR determinant, MHC class II DP determinant, MHC class II DQ determinant, MHC class II

Dakopatts Becton Dickinson Beeton Dickinson Becton Dickinson

BBIG-PI BBIG-I1 BBIG-V~ BBIG-E6

PECAM (CD31) ICAM-1 (CD54) VCAM-1 E-selectin

British Bio-technology British Bio-technoiogy British Bio-teehnology British Bio-teclmology

Hl_,e-1 RFD7 Leu 4 Leu 3a Leu 2a a-IL-2R Ki-67 M708 M752 M851 M877 M852

Leucocyte common antigen (CD45) Macrophages Pan T cells (CD3) T helper/inducer ceils (CD4) T eytotoxic/suppressor (CD8) 1I-2 receptor, a chain (CD2.5) Nuclear antigen proliferating cells Pan B cells (CD22) Neutrophil elastase a-Smooth muscle actin Prolyl 4-hydroxylase (fibroblasts) NCAM, NK cells (CD56)

Becton Dickinson Royal Free Hospital Becton Dickinson Beeton Dickinson Becton Dickinson Becton Dickinson Dakopatts Dakopatts Dakopatts Dakopatts DakQpatts Dakopatts

"Absent from DQw2.

TransplantImmunology 1993; 1:294-301

source

296 PM Taylor et al.

Table 2 Expression of antigens on endothelium from control coronary arteries (n -- 12)

Table 4 Expression of antigens on endothelium from transplanted coronary arteries (n = 6)

MAb specificity

Intensity

Quantity

Score

MAb specificity

Intensity

Quantity

Score

EN4 PaI-E vWF

2-3 2 3

3-4 3 4

6 5 7

EN4 Pal-E vWF

2-3 2 3

3-4 3 4

6

MHC Class I DR antigen DP antigen DQ antigen

2-3 2-3 2 1-2

3-4 3 1-2 1

6 5--6 3--4 2-3

MHC Class I DR antigen DP antigen DQ antigen

2-3 2-3

3-4

6

3 2

5--6 4

1-2

1-2

3

PECAM ICAM-1 VCAM-1 E-Selectin

2-3 2 2 2-3

3--4 3 2 2-3

6 5 4 5

PECAM ICAM-1 VCAM-1 E-Selectin

2-3 2-3 2 2-3

3-4 3 2-3 2-3

6 5-6 4-5 5

Prolyl 4-hydroxylase

1-2 2 3

1-2 3-4 1

3 5-6 4

Prolyl 4-hydroxylase a-Smooth muscle actin CD45

2-3 2 2-3

2-3 3 1-2

5 5 4

a-Smooth muscle actin

CD45

2

5 7

expressing prolyl 4-hydroxylase were found in eight coronaries. Neither B cells nor NK cells were present in control coronaries. The only leucocytes present in the media were macrophages which were found in eight specimens. Medial smooth muscle cells (SMC) expressed MHC class I and a-smooth muscle aetin but not MHC class II or ICAM-1. SMC within the intima also expressed prolyl 4-hydroxylase to varying degrees.

scores shown in Tables 3 and 5 are averages of positive scores only.

Results Control coronary arteries Endothelial cells lining all control coronary arteries strongly expressed EN4, Pal-E, vWF, MHC class I, D R antigen, PECAM, ICAM-1, E-selectin and more variably DP antigen and VCAM-1. DQ antigen was only weakly expressed (Figure 1, Table 2). There was no difference between the expression of donor and recipient coronary arteries. The luminal surface of all coronaries expressed a-smooth muscle aetin which was also expressed by myocardial capillary endothelium in adjacent pieces of tissue and 11/12 weakly expressed prolyl 4-hydroxylase, a fibroblast antigen (Figure 2). Four control coronaries showed minimal disruption of their endothelium. Some leucocytes, usually macrophages, T cells and occasionally neutrophils were present on or under the endothelium of all control coronaries and within the intima of all but one (Figure 1F, Table 3). Cells expressing a-smooth muscle actin were always present within the intima and cells weakly

Transplanted coronary arteries There was no significant difference in the expression of MHC or adhesion molecules on endothelial cells from control arteries and those from transplant patients (Figure 3, Table 4). Twelve transplanted coronary arteries showed varying degrees of disruption of their endothelium and in six cases there was tittle endothelium remaining. In specimens where some endothelium was present (six), it strongly expressed EN4, Pal-E, vWF, MHC class I, D R antigen, PECAM, ICAM-1, E-selectin and to a lesser extent DP antigen and VCAM-1. Again, DQ antigen was only weakly expressed. As we had previously observed in the control group, there was a strong expression of a-smooth muscle aetin on the luminal surface. However, we now also observed a strong expression of prolyl 4-hydroxylase (Figure 2). In specimens where endo-

Table 3 Cells present in control coronary arteries (n = 12) Luminal surface

Intima

Media

MAb specificity

No. +ve

Score

No. +ve

Score

CD45 Macrophage CD3 CD4 CD8

12 8 7 10 5

2 1-2 0-1 1 0-1

11 6 8 8 7

2 1-2 1 1 0-1

9 8 0 7 0

1-2 1 0 1 0

0

0

0

CD22

Neutrophil elastase VCAM-1 ICAM-1 CD25 Ki-67 a-Smooth muscle actin Prolyl 4-hydrox3,1ase

0

4 12 12 0 1 12 11

NA, not applicable.

Transplant Immunology 1993; 1:294-301

0

3 NA NA 0 0-1 8 3-4

0

4 6 6 0 0 12 8

1-2 1-2 1 0 0 NA 1-2

No. +ve

0 5 9 0 0 12 12

Score

0 1 1 O 0 10 5-6

Coronary artery immunogenicity 297

I

-. - ' ~ ' ~ ~

Figure 1 Photomicrograph (immunoperoxidase, X195) of serial 6 ~ ¢ryostat sections showing control coronary artery from explanted recipient heart stained with monodonal antibodies against PECAM (A), ICAM-1 (B), E-seleclln (C), VCAM-1 (D), DR antigen (E) and CIM5 (F). A, B, C and E endothelium strongly stained, D more patchy expression and F endothelium negative, but leucocytes within the tissue positively stained. thelial cells were disrupted, infiltrating leucocytes were responsible for much of the MHC and adhesion molecule expression on the luminal surface.

TransplantImmunology 1993; 1:294-301

More cells were present on or under the luminal surface and within the intima of the transplanted coronary arteries, particularly those with intimitis, than were found in control

298

PM Taylor et al.

Ie~are 2 Photomicrograph (immunoperoxidase, x390) of serial 6 pJn cryostat sections of control coronary artery from harvested donor heart stained with monocional' antibodies against PECAM (A) and prolyl 4-hydroxylase (B) and of transplanted coronary artery stained with monoclonal antibodies against PECAM (C) and prolyl 4-hydroxylase (D). A and C endothelium strongly stained with PECAM. Cells on the luminal surface and within the intima and media show stronger expression of prolyl 4-hydroxylase in D (transplanted coronary artery) than B (control coronary artery).

specimens (Figure 3, Table 3). They were of the same cell type, i.e. macrophages, T cells, prolyl 4-hydroxylase positive cells, or-smooth muscle actin positive cells and sometimes neutrophils. The most striking difference between the transplanted groups was the presence of B cells, IL-2R and Ki-67 (proliferating) cells at the luminal surface or within the intima of four of the six specimens in the intimitis/TX-ACS group. A s well as macrophages, T cells were also found in the media of seven transplanted specimens. PECAM, ICAM-1 and VCAM-1 unlike E-selectin are not restricted to endothelial cells. Thus expression of P E C A M , ICAM-1 and VCAM-1 in the intima and media reflected the presence of macrophages, T cells and SMC. A s had been found in control coronaries, medial SMC were M H C class I, a-smooth muscle actin and prolyl 4-hydroxylase positive but did not express MHC class H or ICAM-1. Again,: we did not find any NK cells in transplanted coronaries.

Transplant Immunology 1993; 1:294-301

Discussion Previous studies from this centre indicated that endothelium from coronary arteries showed strong expression of MHC antigens and adhesion molecules unlike other large vessels such as aorta, pulmonary artery and umbilical vein. Like myocardial capillary endothelium, 8 coronary artery endothelium constitutively expresses MHC class I and class II antigens, P E C A M and ICAM-1, and in addition expresses VCAM-1 and E-selectin. Here we have confirmed and extended these observations using a larger series of coronary arteries. The question arises whether coronary arteries are 'normaUy' immunologically active or whether these molecules have been induced in the specimens used in this study. Our specimens were derived either from donor hearts that were not used or from explanted hearts from patients receiving transplants for conditions not thought to involve the coronaries (cardiomyopathy, congenital heart disease). It is quite possible that cytokines released through intensive care ther-

Coronary artery irnmunogenicity 299

,r

. .

o

¢-~"

~

2"

l

v

%

F

Figure 3 Photomicrograph (immunoperoxidase, X390) of serial 6 ttm cryostat sections of transplanted coronary artery with TX-ACS stained with" monoclonal antibodies against PECAM (A), ICAM-1 (B), E-selectin (C), VCAM-1 (D), DR antigen (E) and CD45 (F). A-E endothelium strongly stained (C, expression restricted to endothelial cells; D, other areas of vessel showed more patchy expression of VCAM-1). F endothelium negative, but infiltrating leucocytes positive. apy (for the donors) caused upregulation in the donor specimens. However, the explanted heart is removed within five minutes of cardiopulmonary bypass, thus it is not possible

TransplantImmunology1993; 1:294-301

to identify the agent causing upregulation in these cases. In any case, the question of 'normality' is academic when considering the aetiology of TX-ACS. The heart donated to

300

PM Taylor et al.

Table 5 Cells present in transplanted coronary arteries Luminal surface MAb speeiticity

No. +ve

Intima Score

No. +ve

Media Score

No. +ve

Score

Without TX-A CS/intimitis (n = 8) CD45 8 Macrophage 8 CD3 8 CD4 6/7 CD8 6 CD22 1/7 Neutrophil elastase 31"/ VCAM-1 6 ICAM-1 8 CD25 1/7 Ki-67 0 a-Smooth muscle actin 8 Prolyl 4-hydroxylase 7/7

4--5 3 1 2-3 0-1 0-1 1 NA NA 0-1 0 8 6

7 6 6 5/7 6 1/7 1/7 3 6 1/6 0 8 6/7

3 2-3 1-2 2-3 1 1 0-1 2 2 1 0 NA 2-3

7 7 3 5/7 2 0 0 4 5 0 1 8 7/7

2 1-2 0-1 1-2 0-1 0 0 1 1 0 1 10 7

With TX-ACS/intimitis (n = 6) CD45 Macrophage CD3 CD4 CD8 CD22 Neutrophil elastase VCAM-1 ICAM-1 CD25 Ki-67 a-Smooth muscle actin Prolyl 4-hydroxylase

5 3-4 2 2-3 1-2 2 1 NA NA 1 1 7 5

6 6 6 6 5

5--6 4-5 2-3 3--4 2

6 6 4 6 4

2 2 1-2 1-2 1

4

2

0

0

2 5 6 4 3 6 6

2 2 4-5 1-2 1 NA 3-4

1 3 6 1 1 6 6

1 2 2 2 1 10 8

6 6 6 6 5 1 2 5 6 4 3 6 6

NA, not appficable. the recipient already has coronary arteries that strongly express donor MHC antigens and other molecules necessary to elicit an immune response. In view of the appearance of 'normal' coronaries it is perhaps not surprising that we failed to find significant differences in MHC or adhesion molecules in the transplanted vessels. This study was, however, made more difficult by loss of endothelium. Whether this loss occurred/n vivo or during sample preparation is impossible to say. We use various markers for identification of endothelial cells and some of these also stain macrophages (EN4, PECAM). Thus, care must be taken in interpreting the results. PaL-E, like E-selectin, is restricted to venule endothelium in the heart, but unlike E-selectin is also expressed by some capillary endothelium.9"1° Both PaL-E and E-selectin proved to be useful markers of coronary artery endothelium. Unlike many investigators, we do not use vWF as a marker of endothelial cells. The stain can diffuse from the endothelial cells into underlying layers and give a false impression. Indeed, we find vessels deliberately stripped of endothelium still express vWF (unpublished observations). It was found useful to divide the transplanted coronaries into those with and those without evidence of intimitisfI'XACS. There were two main differences between the control and transplanted coronaries: the number of cells in the intima and media, which was emphasized in the intimitls/rX-ACS group, and the presence of 'fibroblast-like' cells expressing prolyl 4-hydroxylase. The antibody used to detect fibroblasts is directed against prolyl 4-hydroxylase which is an enzyme present in activated fibroblasts.11,12 We found a stronger expression of this enzyme in the transplanted specimens. The enzyme is also found in SMC, but there are no reports to date Transplant Immunology 1993; 1:294-301

of prolyl 4-hydroxylase being present in endothelial cells. However, it has recently been reported that cultured human endothelial cells form type I collagen fibrils on their surface. 13 These authors suggest that this may be important in the pathogenesis of atherosclerosis. Prolyl 4-hydroxylase was the only antigen that was upregulated in the transplanted coronaries. Thus, the presence of these 'fibroblast-like' cells on the luminal sudace may play an important role in the pathogenesis of TX-ACS. We found increased numbers of macrophages, T cells, B cells, cells expressing IL-2 receptor, cells expressing Ki-67 and cells expressing the fibroblast antigen in the intima of the transplanted coronaries, especially those with TX-ACS and/ or intimitis. The majority of these cells were most often found just under the endothelium, as has been previously observed. 14'x5 Like Salomon et aL, TM but unlike Hruban et al., ~4 who suggested there was a predominance of CD8 positive T cells based on one specimen; we found equal numbers of CD4 and CD8 positive T cells. These cells can release cytokines which amongst other effects will upregulate MHC and adhesion antigen expression, perhaps even increasing their affinity for their ligands, allowing more cells into the intima. In these studies medial SMC did not express MHC class II or ICAM-1. However, it would appear that once these SMC had migrated from the media into the intima they did express these antigens. Only a few macrophages were found in the media from control coronaries, whereas T cells were also found in the media of about half of the transplanted coronaries, particularly those with TX-ACS and/or intimitis. The presence of T cells in the media of transplanted coronaries with ACS has been previously reported. 14 It would appear that coronary arteries are highly im-

Coronary artery imrnunogenicity 301

munogenic prior to transplantation, strongly expressing MHC (class I and class II) and adhesion molecules (PECAM, ICAM-1, VCAM-1 and E-seleetin). This suggests the endothelium would need little, if any, stimuli to initiate as well as be a target for an immune response ultimately leading to TXACS. Factors such as MHC mismatch, a high frequency of precursor cytotoxic T cells (McDouall, Rose and Yacoub, in preparation) and anti-endothelial antibodies5 may predispose the transplant recipient to the disease.

Acknowledgement Source of funding: British Heart Foundation.

References 1 Banner NR, Fitzgerald M, Khaghani A et at Cardiac transplantation at Harefield Hospital. In: Terasaki P ed. Clinical transplants. Los Angeles: UCLA Tissue Typing Laboratory, 1987: 17-26. 2 Utresky BF, Murali S, Reddy PS eta/. Development of coronary artery disease in cardiac transplant patients receiving immunosuppressive therapy with cyciosporine and prednisone. Orculation 1987; 76: 827-34. 3 Gao S, Hunt SA, Sehroeder JS. Accelerated transplant coronary artery disease. In: Yacoub MH, Loop FD eds. Seminars in thoracic and cardiovascular surgery. Philadelphia: Saunders, 1990: 241-49. 4 Ross R. The pathogenesis of atherosclerosis: an update. N EnglJ Med 1986; 314: 488-500. 5 Dunn MJ, Crisp SJ, Rose ML et al. Anti-endothelial antibodies

Transplant I m m u n o l o g y 1993; 1:294-301

and coronary disease after cardiac transplantation. Lancet 1992; 339:. 1566-70. 6 Springer TA. Adhesion receptors of the immune system. Nature 1990; 346: 425-34. 7 Pober JS, Cotran RS. The role of endothelial cells in inflammation. Transplantation 1990,50: 537--44. 8 Page C, Rose ML, Yacoub MH st al Antigenic heterogeneity of vascular endothefium. A m J Patho11992; 141: 673--83. 9 Hengstenberg C, Rose ML, Page C et aL Immunoeytochemical changes suggestive of damage to endothefial cells during rejection of human cardiac allogratts. Transplantation 1990;49: 895-99. 10 Taylor PM, Rose ML, Yacoub MH st aL Induction of vascular adhesion molecules during rejection of human cardiac allografts. Transplantation 1992;54: 451-57. 11 Konttinen YT, Nykanen P, Nordstrom D et aL DNA synthesis in prolyl 4-hydroxylase positive fibroblasts in situ in synovial tissue. An autoradiography--immunoperoxidase double labeling study. J Rheumato11989; 16: 339--45. 12 Janin A, Konttinen YT, Gronblad M st al. Fibroblast markers in labial salivary gland biopsies in progressive systemic sclerosis. Clin Exp Rheumato11990; 8: 237-42. 13 Yamamoto K, Yamamoto M, Ooyama T et al. Type I collagen fibril formation by human vascular endothelial cells in culture. Artery 1992; 19: 112-22. 14 Hruban RH, Beschomer WE, Banmgartner WA et al. Accelerated arteriosclerosis in heart transplant recipients is assodated with a T-lymphocyte-mediated endothelialitis. A m J Pathol 1990; 137: 871-82. 15 Saiomon RN, Hughes CCW, Schoen FJ st al. Human coronary transplantation-associated arteriosclerosis: evidence for a chronic immune reaction to activated graft endothelial cells. A m J Pathol 1991; 138: 791-98.