Secondary Logo

Share this article on:

CELL AND EXTRACELLULAR MATRIX REJECTION IN ARTERIAL CONCORDANT AND DISCORDANT XENOGRAFTS IN THE RAT1

Allaire, Eric2,3; Mandet, Chantal4; Bruneval, Patrick4; Bensenane, Soraya3; Becquemin, Jean-Pierre2; Michel, Jean-Baptiste3

Immunobiology

Vascularized xenografts are rejected acutely and hyperacutely in concordant or discordant combinations, respectively. We investigated the impact of the donor-recipient combination on the rejection of arterial xenografts, analyzing the cellular and extracellular matricial compartments.

Aortic xenografts were performed in a concordant (hamster) and a discordant (guinea pig) combination with Lewis rat. Graft cells and immune effectors were characterized by immunohistochemistry after 15 min and up to 30 days postimplantation. Macroscopic and microscopic structure of the grafts was studied at 60 days.

IgG in the concordant combination and C3, C5b9, and IgM in the discordant combination deposited on endothelial cells, acutely and hyperacutely, respectively. The same immune effectors deposited on medial smooth muscle cells, but later than on endothelial cells.

In both combinations the medial extracellular matrix was covered by IgM and IgG and infiltrated by monocytes (90%) and T lymphocytes (10%), with elastinolysis in the vicinity of monocytes. However, elastin resorption in the media at day 60 differed in concordant and discordant xenografts(75±10% and 99±1%, respectively). Intimal thickening and aneurysm developed in concordant and discordant combinations, respectively.

Unlike arterial allografts, arterial xenografts are not a homogeneous group. The donor-recipient combination determines the mechanism and the timing of graft cell rejection, as well as the magnitude of medial elastin injury. As a consequence, chronic graft remodeling differs in the two combinations.

Unit 367 INSERM and Unit 430 INSERM, Paris; and Service de Chirurgie Vasculaire, Hopital Henri Mondor, Creteil, France

1This study was supported by research grants from the Fondation de l'Avenir, and from CNAMT/INSERM, Paris, France.

2 Service de Chirurgie Vasculaire, Hopital Henri Mondor, 51 Avenue du Marechal de Lattre de Tassigny, 94010 Creteil Cedex, France.

3 Unit 367 INSERM. 17 rue du Fera Moulin, Paris.

4 Unit 430 INSERM, Hopital Broussais, Paris.

Received 17 January 1996.

Accepted 17 April 1996.

Aortic engraftment in rat is extensively used to model arterial allograft rejection (1-6), and rodent xenograft combinations are used to investigate the mechanisms of xenograft rejection (7). Early reports of arteries following long term implantation described more severe injury in xenografts than in allografts, with a high frequency of aneurysmal dilation(8). Nevertheless the mechanisms of arterial xenografts rejection are unknown.

Xenograft rejection has been characterized in heart and other vascularized organs (9, 10). Xenografts are referred to as“concordant” and “discordant” according to the timing and to the immune effectors depositing on graft capillary endothelial cells. Concordant xenografts are rejected within days, when immunoglobulins specific to the graft have been induced by the presence of the graft, although at the time of engraftment immunoglobulins able to recognize the graft are present in the serum of the recipient (11-13). Discordant xenografts are rejected within minutes by effectors present in the serum of the recipient at the time of engraftment(14, 15). The concordant and discordant character of vascularized xenografts is mainly determined by the donor-recipient combination and not by which organ is grafted (9, 10).

The studies of allograft rejection have shown that the mechanisms of arterial and vascularized graft rejection are not similar(1-6). The fact that arteries are made of a dense extracellular matrix bearing their elementary mechanical properties might account for this difference. In a previous work, we showed that guinea pig aortas grafted into rats become aneurysmal with an intense destruction of the extracellular matrix in the media, whereas aortic allografts do not become aneurysmal and have their medial extracellular matrix relatively preserved (16).

The fate of the extracellular matrix has never been investigated in xenografts. The reason could be that in vascularized xenografts organ dysfunction is primarily due to graft cell injury. However, in the aorta the extracellular matrix is an abundant material supporting blood pressure-induced tensional stress. Its targeting and injury as a xenogenic structure should determine both the mechanism of rejection and the remodeling of the arterial wall. The relationship between extracellular matrix injury and the remodeling of arteries is suggested by the pathophysiology of aneurysms(16-18).

We hypothesized that graft cells and extracellular matrix both play a role in arterial xenograft rejection. The aim of this work was to assess the effectors, the timing and the magnitude of arterial cell and extracellular matrix rejection in arterial xenografts, and the effect of the donor-recipient combination on these parameters. Specific questions were addressed. Are cells in arterial xenografts rejected by the same effectors as endothelial cells in vascularized xenografts? Are cells other than endothelial cells targeted by the immune system in arterial xenografts? Does the donor-recipient combination influence the rejection of the arterial extracellular matrix, and hence the chronic remodeling of arterial xenografts?

To investigate these questions, we performed aortic engraftments in two murin donor-recipient combinations referred to as concordant (hamster into rat) and discordant (guinea pig into rat) (9, 10).

Back to Top | Article Outline

MATERIALS AND METHODS

Animal and Graft Procurement

Recipients were male inbred Lewis rats (Iffa-Credo, Lyon, France, body weight 200-250 g). Graft donors were male outbread guinea pigs (Centre Ardenay, France, body weight 350 to 400 g) and male outbred hamsters (Centre Ardenay, France, body weight 90-100 g) and Lewis rats for isografts.

Species were chosen from known concordant (hamster) and discordant (guinea pig) xenograft combinations with rat for vascularized organs(10). To match the size of the graft with the recipient aorta, we used hamster thoracic aortas, and guinea pig and rat abdominal aortas. Animal care complied with the European Community Standards(Ministère de l'Agriculture, France, authorization No. 00577).

Back to Top | Article Outline

Experimental Design

The rejection was studied sequentially in the 2 xenograft combinations and in isografts. A total of 5 xenografts in each combination and 4 isografts were studied at the following time points: 15 min; 1, 6, 12, 24, and 48 hr; and 5, 8, 15, 30, and 30 days after engraftment. These grafts were used as follows: 2 xenografts in each combination and 2 isografts were used for standard histology. Three xenografts in each combination and 2 isografts were used for immunohistochemistry. Ten concordant and 10 discordant xenografts and 10 isografts were studied at 60 days.

Back to Top | Article Outline

Graft Transplantation, Harvest, and Preparation

Donors and recipients were operated on simultaneously under anesthesia with pentobarbital (5 mg/100 g body weight i.p.), with an operating microscope. A 10 mm segment of the recipient aorta was removed from below the renal arteries and above the aortic bifurcation, and replaced by an equal length of graft(19). After surgery, rats were fed with standard diet and given no medication.

For graft harvest, animals were anesthetized with pentobarbital, and killed by an overdose of pentobarbital i.v. Some grafts were fixed in situ by perfusion of formalin at 100 mmHg for 30 min through a catheter introduced into the left iliac artery. These grafts were paraffin embedded and cut in longitudinal sections. Other grafts were rinsed with 9% NaCl, embedded in OCT compound (Tissue-Tek, Miles Inc.), and snap frozen into liquid nitrogen. Cross-sections from the center of these grafts were made with a cryostat (5 mm thickness) and kept at -70 °C until use.

Back to Top | Article Outline

Microscopic Structure of the Graft

For standard histology, formalin-fixed longitudinal sections and cryostatic cross-sections from grafts were stained with Verhoeff stain for general structure, elastin, and collagen.

Back to Top | Article Outline

Immunohistochemistry

The immune effectors and graft cells were identified by immunostaining techniques. Immunostaining of monocyte-macrophages and smooth muscle cells was also performed on longitudinal sections from the formalin-fixed grafts harvested at 60 days. The antibodies used are listed inTable 1.

A direct immunofluorescence technique was used for rat IgG, C3 andα-actin. Immunohistochemistry used for the other antibodies was an alkaline phosphatase-anti-alkaline phosphatase (APAAP) technique counterstained with hematoxylin. After acetone fixation, sections were exposed to the primary monoclonal antibody diluted in 0.5% bovine serum albumin, followed by a polyclonal rabbit anti-mouse IgG (Dakopatts Laboratories, Denmark) diluted in 10% normal rat serum to avoid nonspecific binding, and by mouse APAAP complexes (Dakopatts Laboratories). Sites of alkaline phosphatase fixation were identified by incubation with a fast red substrate system(Dakopatts Laboratories). The absence of nonspecific labeling was assessed after omission of the primary antibody, by incubation with nonimmune sera from rabbit and mice instead of primary antibody (Cappel Products). In addition, the antihuman PECAM antibody (IgG, dilutions 1/10 to 1/100) failed to stain any tissue in the study except hamster endothelial cells (see below), and was used as an irrelevant antibody as another negative control.

Back to Top | Article Outline

Immunohistochemistry-Labeling Semiquantification

A grid eye-piece on a microscope was used to count the cells labeled on two cross-sections for each graft. The grid was moved at regular intervals at a 40× magnification. The counts were made for each of the 3 areas of the graft: intima, media, and adventitia. The results were averaged for each compartment and expressed as the number of cells per 5.3 square-millimeter.

Back to Top | Article Outline

Analysis of Graft Cell and Extracellular Matrix Injury

The impact of the donor-recipient combination on arterial injury was assessed based on the timing of graft cell disappearance, on the graft diameter and microscopic aspect, and on quantification of elastin, indicating the extend of the injury of the extracellular matrix.

Back to Top | Article Outline

Time of Graft Cell Disappearance

Two isografts and 3 xenografts in each combination were harvested 15 min; 1, 6, 12, 24, and 48 hr; and 5, 8, 15, and 30 days after engraftment. The presence of endothelial cells on the intima and smooth muscle cells in the media was determined. The time of rejection was defined as the time after engraftment when all endothelial cells or smooth muscle cells had disappeared on frozen cross-sections. Because hamster to rat grafts were infiltrated by inflammatory cells at the time of cell rejection, arterial wall cells were identified with specific monoclonal antibodies (Table 1). Hamster smooth muscle cells were identified by the anti-rat α-actin antibody. On control sections, the anti-human PECAM antibody electively stained hamster endothelial cells. No other cell type in the hamster arterial wall, or rat inflammatory cells in the grafts or on rat spleen frozen sections were stained by the antibody. Also the antihuman PECAM antibody was used to identify hamster endothelial cells. Guinea pig endothelial and smooth muscle cells were identified by a hematoxylin nuvlear staining.

Back to Top | Article Outline

Measurement of Graft Diameter

The width of 5 grafts in each group was measured with a grid in the microscope eyepiece, immediately after transplantation (D0), and before euthanasia 60 days after transplantation (D60). The diameter increase for each graft was calculated as follows: D diameter = (D60-D0)/D0×100.

Back to Top | Article Outline

Quantification of Elastin in the Media

The medial elastin content was measured on orcein-stained longitudinal sections from 5 grafts harvested 60 days after engraftment in each group, and from 5 native Lewis rat abdominal aortas, 5 hamster thoracic aortas, and 5 guinea pig abdominal aortas. The grafts and the native aortas had been fixed in situ with formalin before harvest. The quantification was made by a computer-assisted morphometric technique (NS 15000; Nachet-Vision, Paris, France) (20). The measured parameters were: (1) the relative area occupied by the elastic network (S%), and (2) the mean thickness of the media (tm). Measurements and calculations were performed in 10 regularly spaced microscopic fields in each section and averaged for each graft. The medial elastin content (MEC) was calculated as follows: MEC = S% x tm, for each graft. The mean medial elastic content of native aortas (natMEC) was calculated in each group. The mean medial elastin resorption ratio for each graft was: elastin resorption = (MEC-natMEC)/natMEC.

Back to Top | Article Outline

Statistical Analysis

Quantitative results were expressed as means ± SD Comparisons of diameters were done by analysis of variance (ANOVA), followed by the ScheffeF test for comparisons between 2 groups. The numbers of ED1 positive cells infiltrating the adventitia were pooled in two groups, isografts versus xenografts, and compared at each time-point with the nonparametric Mann-Whitney U test. The Mann-Whitney test was also used to compare parameters between the 2 xenograft combinations. A probability of <0.05 was accepted as significant.

Back to Top | Article Outline

RESULTS

Arterial Graft Cell and Extracellular Matrix Injury in Concordant and Discordant Xenograft Combination

Timing of graft cell rejection (Table 2). At 1 hr after engraftment in the 3 groups, endothelial cells covered most of the intima, and smooth muscle cells were present in the media. Endothelial and smooth muscle cells in isografts remained intact at later time points.

The endothelial cells disappeared between day 5 and day 8 in the concordant combination, and between 6 and 12 hr in the discordant combination. Smooth muscle cells disappeared later than endothelial cells in xenografts, between day 15 and day 30 in the concordant combination, and between 12 and 24 hr in the discordant combination.

Smooth muscle cells stained by the anti-α-actin antibody in the concordant combination disappeared at the leading edge of a monocytic infiltrate penetrating into the media. In the discordant combination, medial smooth muscle cells had disappeared before any inflammatory cell had marginated on the intima or penetrated into the media of the grafts.

Graft diameter (Table 3). No graft ruptured or occluded. Sixty days after engraftment, the 10 grafts in the concordant combination were all slightly dilated (D diameter = 31.7±14.7%), whereas the 10 grafts in the discordant combination were constantly aneurysmal(D diameter = 329.6±65.4%). The difference between the groups was significant (the Scheffe F test: 148.0, P<0.0001).

Graft microscopic changes. The native aortas in the 3 species, as well as the isografts at the different time-points after transplantation, had an unthickened intima, an internal elastic lamina, and parallel elastic lamelae in the media, and an adventitia mainly composed of collagen(Fig. 1, panel A to C). However, the elastic network was looser in the guinea pig aorta.

Intima. In the concordant combination, an intimal thickening was present at day 30 and day 60 (Fig. 1, panel D and E). At day 60, it contained few elastic fibers (Fig. 1, panel F), collagen, α-actin positive cells (Fig. 1, panel F) and few monocyte-macrophages (Fig. 1, panel D).

In the discordant combination, a thrombus appeared between day 8 and 15 on the intima, and thickened as the graft dilated. At day 60, it appeared as an organized collagen-rich fibrosis with no elastic fibers and no α-actin positive cells (Fig. 1, panel G to I).

Media. In both types of xenografts, elastin fibers in the media were destroyed (Fig. 1, panel D and G) by invading monocyte-macrophages (Fig. 2). Degradation was apparent at day 30 in the concordant combination and at day 15 in the discordant combination. At 60 days multinucleated giant cells were in contact with the remaining elastic lamelae, with figures of elastinolysis. The areas of destroyed media were replaced by a dense collagen fibrosis.

Since sequential study showed no neoelastin deposition stained with orcein in the media after the lysis of the native elastin fibers, elastic loss was quantified in xenografts. At 60 days after engraftment, the elastic resorption was 75±10% in the concordant combination and 99±1% in the discordant combination. This difference was statistically significant (results from 5 grafts in each group, Mann-Whitney: Z2.6,P=0.009).

Adventitia. In the 3 types of grafts at day 60, a collagen-rich fibrosis had developed in the adventitia. In xenografts, the fibrosis replacing resorbed grafted media could not be distinguished from the adventitial fibrosis (Fig. 1, panel G to I).

Back to Top | Article Outline

Immunolabeling of the Effectors of Graft Cell and Extracellular Matrix Rejection

Humoral effectors (Table 2). For all humoral effectors (complement, IgM, IgG) the staining on isografts was negative. In both types of xenografts, humoral effectors deposited on endothelial cells earlier than on medial smooth muscle cells.

Complement. In the concordant combination endothelial and smooth muscle cells were negative with anti-C3 and anti-C5b9 antibody(Fig. 3, panel A). C5b9 deposits in this group occurred only in the media from day 8 to 30 and were interpreted as deposits of circulating complexes on the acellular injured media(21). In the discordant combination, anti-C3 antibody staining was intense after 15 min on endothelial cells, and after 1 hr on medial smooth muscle cells. The anti-C5b9 antibody staining on endothelial and medial smooth muscle cells was negative at 1 hr, and was positive at 6 hr(Figure 3, panel B). At day 15 in the discordant combination, medial remnants were positive with the anti-C5b9 antibody.

In the concordant combination, the intima was positive for IgG at day 8. The media was negative for IgM (Fig. 3, panel C) and IgG up to day 8. At day 15 and 30 the media was positive for IgM and IgG in the first subluminal interlamelar space where smooth muscle cells had disappeared(Fig. 4, panel A, B, and C). The whole media was positive for IgM and IgG at day 30.

In the discordant combination, endothelial cells were positive for IgM from 15 mn to 6 hr and the medial smooth muscle cells at 6 hr(Figure 3, panel D). At these time points, the IgG staining was negative. The whole media was positive for IgM and IgG at day 30.

Back to Top | Article Outline

Cellular Effectors

Adventitia. Inflammatory cells accumulated on the intima and in the adventitia of the aortic wall. This accumulation displayed a three-stage schedule which was observed in the two types of xenografts. However, each phase started and was full-blown earlier in the discordant than in the concordant combination.

The first stage was recruitment of inflammatory cells in the adventitial space of iso- and xenografts. The adventitial space was the first area where inflammatory cells accumulated. The schedule was similar in the three types of grafts. Polymorphonuclears (90% of total inflammatory cells), T lymphocytes, and ED1+ monocyte-macrophages accumulated as soon as 1 hr after engraftment; B cells at 6 hr; and ED2+ macrophages at 24 hr. After 24 hr, polymorphonuclears disappeared.

The second stage was the xenograft-specific phase of intimal and adventitial inflammatory infiltrate differentiation. The second stage started 24 hr after engraftment in xenografts only (Fig. 5). The number of ED1+ monocyte-macrophages increased in the adventitia of xenografts, but not of isografts (P<0.05, Mann-Whitney U test, for ED1+ cells at each time point after 6 hr). Cells in the adventitia of xenografts were activated (increased anti-ICAM-1 and anti-LFA-1 stainings).

In both types of xenografts, T lymphocytes (CD4/CD8=2) appeared at day 8, and had vanished at day 15 in the adventitia. ED1+ monocyte-macrophage started to accumulate later in the concordant combination (day 5) than in the discordant combination (48 hr). At 48 hr, the number of adventitial ED1+ cells was lower in the concordant combination (29.3±6.5) than in the discordant combination (42.0±14.4) (analysis of 3 grafts in each group, Mann-Whitney: Z = -1.96, P=0.049).

The third stage was decrease of inflammatory infiltrate in the adventitia and increase in the intima. In the discordant combination, the decrease of adventitial inflammatory infiltrate followed the disappearance of the medial extracellular matrix at day 30. At this time point, the number of adventitial ED1+ cells was higher in the concordant combination (61.7±12.4) than in the discordant combination (10.0±1.8) (analysis of 3 grafts in each group, Mann-Whitney: Z = -1.96, P=0.049).

Media. No inflammatory cell penetrated into the media of isografts. In both types of xenografts, inflammatory cells located in the adventitia at day 8 had penetrated into the media at day 15(Fig. 2). The penetration of inflammatory cells into the media differed in the concordant and the discordant combinations. At day 15, the inflammatory cells had penetrated more deeply into the media in the discordant than in the concordant combination. The number of cells positively stained for the activation markers, ICAM-1 and LFA-1α, was maximum at day 30 in the concordant and at day 15 in the discordant combination. At day 30, activation marker staining and the number of ED1+ macrophages had decreased in the completely destroyed media of discordant, and had increased in the incompletely destroyed media of concordant xenografts. In both combinations infiltrating cells were ED1+/ED2-monocytes (90%), and T lymphocytes (10%,CD4/CD8=1.5). In both groups, T lymphocytes (CD4/CD8=0.75) appeared in the fibrous scar tissue replacing the destroyed media.

Back to Top | Article Outline

DISCUSSION

Since the function of the arterial wall and the viability of the graft cells are dissociated, we are able to analyze independently the cells and the extracellular matrix during a long course of rejection with no immunosuppressive treatment in concordant and discordant combinations. We show that other cells than endothelial cells are rejected in arterial xenografts.

We show that graft cells are rejected differently in the concordant and discordant combinations. As in vascularized organ xenografts, this difference appears in the chronology of events and in the effectors involved. Moreover, the effectors and the chronology of endothelium rejection in arterial xenografts are similar to those of capillary endothelial cell rejection of vascularized xenografts in rat (22). The disappearance of arterial graft endothelial cells in the two combinations correlates temporally with the deposition of the immune effectors. Therefore, a parallel can be drawn between arterialized organ rejection and arterial rejection in the two xenograft combinations, referring to these combinations as concordant and discordant for arterial rejection.

In concordant and discordant arterial xenografts, smooth muscle cells in the media are rejected by the same effectors as endothelial cells, but later. In vascularized xenografts capillary rejection coincides with the onset of graft dysfunction, when humoral effector deposition-IgG in concordant and C3 and IgM in discordant xenografts-is limited to the endothelium(10, 14, 15). Because aortic xenografts do not thrombose, medial smooth muscle cells are subjected to immune effectors from the blood long after endothelial cells have been injured. We report the first in vivo evidence that humoral effectors in both xenograft combinations are able to target cells other than endothelial cells in a model without immunosuppression. Moreover, we confirm that the completion of complement activation from C3 to C5b9 takes more than 1 hr on both endothelial and smooth muscle cells and is probably involved in cell disappearance in discordant arterial xenografts (23).

In the first 24 hr after engraftment, inflammatory cells accumulate in the adventitia in all groups of arterial xenografts. This first stage is probably related to the surgical procedure as shown by the decrease in adventitial inflammatory cells in isografts at later time points. The up-regulation of adventitial inflammation at 24 hr and later is specific to xenografts. A feature of arterial xenograft rejection is the penetration of inflammatory cells into the media in sharp contrast to the low medial penetration of inflammatory cells during arterial allograft rejection in rat(1-6). The localization of the inflammatory infiltrate in the media probably accounts for the severity of the extracellular matrix injury in arterial xenografts, and its functional counterpart, xenograft dilation.

The same inflammatory cell types infiltrate the media in the two xenograft combinations. Activated mononuclear phagocytes synthesize proteolytic enzymes and degrade extracellular matrix proteins, including elastin(24-27). In both xenograft combinations, medial extracellular matrix degradation is a long-lasting process compared with the rejection of the cells. The presence of multinucleated giant cells at the surface of the elastic structures in both xenograft combinations, as well as the high level of expression of LFA-1 integrin, indicate a prolonged stimulation of phagocytic cells by persistent antigens from the extracellular matrix (28, 29). Hence we suggest that the massive and long-lasting character of the inflammatory infiltration of the media during xenograft rejection is due to the chronic presentation of xenoantigens carried by the extracellular matrix, and is a consequence of the immunogenicity of the extracellular matrix in arterial xenografts. Indeed, medial inflammatory infiltration of large arteries is difficult to induce experimentally (18). The passive transfer of T lymphocytes having an autoimmune specificity against vascular smooth muscle cells in mice results in a vasculitis of arterioles and veinules, but spares large elastic arteries (30). In our model three facts suggest that the infiltration of the media could be determined by the immunogenicity of the extracellular matrix: (1) the nonimmunogenic media of arterial iso- and allograft is at best poorly infiltrated (16); (2) the penetration of inflammatory cells into the media of xenografts is associated with the deposition of IgM and IgG on the extracellular matrix; and (3) monocytes disappear from the media of xenografts when the elastic network is completely lysed. The presence of monocytes in the media of xenografts when the injury of the extracellular matrix develops is associated with the deposition of immunoglobulins. Immunoglobulins could recruit monocytes and enhance their proteolytic phenotype.

Medial elastin resorption and its functional counterpart, the dilation of the graft, are both higher in discordant than in concordant xenografts. The difference in the magnitude of extracellular matrix injury in the two combinations could be due to differences in cell rejection, or to differences in extracellular matrix immunogenicity. The phylogenetic distance between guinea pig and rat is greater than between hamster and rat(7). Primary amino acid sequences of elastin and collagen are more divergent between more disparate species(31, 32). As a consequence the extracellular matrix proteins should carry a higher antigenic disparity in discordant than in concordant xenograft combinations (7).

One factor that correlates with the magnitude of extracellular matrix injury and graft dilation in our study is the rapidity and the extent of the infiltration of the media by inflammatory cells. A relatively slow and limited invasion of the media in the concordant xenografts preserves long segments of internal elastic lamina and is associated with an intimal thickening similar to the intimal thickening of arterial allografts in rat(4, 5). This intimal thickening is composed of smooth muscle cells probably from the recipient since hamster smooth muscle cells have been destroyed in the media. A more rapid and/or severe medial injury is associated with aneurysmal formation. In humans, arterial occlusive and aneurysmal diseases are associated with different magnitudes of medial injury. In arterial occlusive disease most of the medial extracellular matrix is preserved although the media is thinned (33). In aneurysms, the medial extracellular matrix is extensively destroyed by phagocytic cells, with a fibrotic reconstruction (also evidenced in discordant xenografts) unable to durably withstand the hemodynamic stress(17, 34-37). In our model, the intensity of the medial extracellular matrix injury is modulated by the donor-recipient phylogenic distance. The structural response of the graft to injury is not a continuum. Our model emphasizes that the magnitude and the time course of the medial extracellular matrix injury switches the evolution of the diseased artery toward an occlusive or an aneurysmal form.

In conclusion, we have shown that arterial xenografts, unlike allografts, are not an homogeneous group. The donor-recipient combination affects the timing and the effectors of the arterial wall cell rejection and the magnitude of medial elastin injury. One common and specific feature of both concordant and discordant xenografts is the massive infiltration of the media by monocytes and T lymphocytes and the deposition of immunoglobulins on the medial extracellular matrix. This infiltration is probably elicited by the immunogenicity of the xenogeneic extracellular matrix. In a chronic stage the heterogenicity of arterial xenografts is expressed by two types of graft remodeling: moderate enlargement and intimal thickening in concordant xenografts, and aneurysmal dilation in discordant xenografts.

Acknowledgments. We are grateful to Drs. H. Mathevon and H. Gstach from the C.H. of Dunkerque and to Pr. Gautier Benoit from the CHR of Lens, France, for their encouragement and support during this study. We thank Dr. B. Levy and Mrs. M. Duriez from INSERM U 141, Paris, France, for their logistic aid for computerized morphometry, and Dr. Couser, Seattle, WA for his help in analyzing the data. We also thank Jean McAllister from the Department of Surgery, University of Washington, Seattle, WA, for her constructive review of the manuscript.

Figure 1

Figure 1

Figure 2

Figure 2

Figure 3

Figure 3

Figure 4

Figure 4

Figure 5

Figure 5

Back to Top | Article Outline

REFERENCES

1. Schmitz-Rixen T, Megerman J, Colvin RB, Williams Am, Abbot WM. Immunosuppressive treatment of aortic allografts. J Vasc Surg 1988; 7: 82.
2. Plissonnier D, Amichot G, Duriez M, Legagneux J, Levy BI, Michel JB. Effect of converting enzyme inhibition on allograft-induced arterial injury. Hypertension 1991; 18: 47.
3. Plissonnier D, Amichot G, Lecagneux J, Duriez M, Gentric D, Michel JB. Additive and synergistic effects of a low-molecularweight, heparin-like molecule and low dose of cyclosporin in the preventing arterial graft rejection in rats. Arterioscler Thromb 1993; 13: 112.
4. Plissonnier D, Nochy D, Poncet P, et al. Sequential immunological targeting of chronic experimental arterial allograft. Transplantation 1995; 60: 414.
5. Isik FF, McDonald TO, Ferguson M, Yamanaka E, Gordon D. Transplant arteriosclerosis in a rat aortic model. Am J Pathol 1992; 141: 1139.
6. Hayry P, Isoniemi H, Yilmaz S, et al. Chronic allograft rejection. Immunol Rev 1993; 134: 33.
7. Chaline J, Cardoso J, Houssin D. Organ xenografting between rodents: an evolutionary perspective. Transplant Int 1994; 7: 216.
8. Carrel A. Ultimate results of aortic transplantation. J Exp Med 1912; 9: 389.
9. Calne RY. Organ transplantation between widely disparate species. Transplant Proc 1970; 2: 550.
10. Auchincloss H. Xenogeneic transplantation: a review. Transplantation 1988; 46: 1.
11. Cosenza CA, Cramer DV, Tuso PJ, Chapman FA, Wang HK, Makowka L. Combination therapy with brequinar sodium and cyclosporine synergistically prolongs hamster-to-rat cardiac xenograft survival. J Heart Lung Transplant 1994; 13: 489.
12. Steinbruchel DA, Nielsen B, Salomon S, Kemp E. Sequential, morphological, and antidonor antibody analysis in a hamster-to-rat heart transplantation model. Transplant Int 1992; 5: 38.
13. Aksentijevich I, Sachs DH, Sykes M. Natural antibodies against bone marrow cells of a concordant xenogeneic species. J Immunol 1991; 147: 79.
14. Gambier L, Salame E, Chereau C, Houssin D. The role of natural IgM in the hyperacute rejection of discordant heart xenografts. Transplantation 1992; 54: 577.
15. Miyagawa S, Horose H, Shirakura R, et al. The mechanism of discordant rejection. Transplantation 1988; 46: 825.
16. Allaire E, Guettier C, Bruneval P, Plissonnier D, Michel JB. Cell-free arterial grafts: morphologic characteristics of aortic isografts, allografts, and xenografts in rats. J Vasc Surg 1994; 19: 446.
17. Dobrin PB, Baker WH, Gley WC. Elastolytic and collagenolytic studies of arteries. Implications for the mechanical properties of aneurysms. Arch Surg 1984; 119: 405.
18. Anidjar S, Salzmann JL, Gentric D, Lagneau P, Camilleri JP, Michel JB. Elastase-induced experimental aneurysm in rat. Circulation 1990; 82: 973.
19. Schwartz CJ, Mitchell JRA. Cellular infiltration in the human arterial adventitia associated with atherosclerotic plaques. Circulation 1962; 26: 73.
20. Levy BI, Michel JB, Salzmann JL, et al. Effects of chronic inhibition of converting enzyme on mechanical and structural properties of arteries in rat renovascular hypertension. Circ Res 1988; 63: 227.
21. Bariety J, Hinglais N, Bhakdi S, Mandet C, Rouchon M, Kazatchkine M. Immunohistochemical study of complement S protein (vitronectin) in normal and diseased human kidneys: relationship to neoantigens of C5b9 terminal complex. Cin Exp Immunol 1989; 75: 76.
22. Leventhal JR, Matas AJ, Sun LH, et al. The immunopathology of cardiac xenograft rejection in the guinea pig-to-rat model. Transplantation 1993; 56: 1.
23. Vakeva A, Morgan BP, Tikkanen I, Helin K, Laurila P, Meri S. Time course of complement activation and inhibitor expression after ischemic injury of rat myocardium. Am J Pathol 1994; 144: 1357.
24. Werb Z, Banda MJ, Jones PA. Degradation of connective tissue matrices by macrophages. I. Proteolysis of elastin, glycoproteins, and collagen by proteinases isolated from macrophages. J Exp Med 1980; 152: 1340.
25. Werb Z, Bainton DF, Jones PA. Degradation of connective tissue matrices by macrophages: III. Morphological and biochemical studies on extracellular, pericellular, and intracellular events in matrix proteolysis by macrophages in culture. J Exp Med 1980; 152: 1537.
26. Shapiro SD, Campbell EJ, Welgus HG, Senior RM. Elastin degradation by mononuclear phagocytes. Proc Natl Acad Sci USA 1991; 624: 69.
27. Katsuda S, Okada Y, Imai K, Nakanishi I. Matrix metalloproteinase-9 (92-kd gelatinase/type IV collagenase equals gelatinase B) can degrade arterial elastin. Am J Pathol 1994; 145: 1208.
28. Papajiannis SP, Spina M, Gotte L. Sequential degradation and phagocytosis of heterologous elastin. Arch Pathol 1970; 89: 434.
29. Falzoni S, Munerati M, Ferrari D, Spisani S, Moretti S, Di Virgilio F. The purinergic P2Z receptor of human macrophage cells. Characterization and possible physiological role. J Clin Invest 1995; 95: 1207.
30. Hart MN, Tassel SK, Sadewasser KL, Schelper RL, Moore SA. Autoimmune vasculitis resulting from in vitro immunization of lymphocytes to smooth muscle. Am J Pathol 1985; 119: 448.
31. Starcher BC, Galione MJ. Purification and comparison of elastins from different animal species. Ann Biochem 1976; 441.
32. Boyd C, Christiano Am, Pierce RA, Stolle CA, Deak SB. Mammalian tropoelastin: multiple domains of the protein define an evolutionarily divergent amino acid sequence. Matrix 1991; 11: 235.
33. Glagov S. Morphology of collagen and elastin fibers in atherosclerotic lesions. Adv Exp Med Biol 1977; 82: 767.
34. Menashi S, Campa JS, Greenhalgh RM, Powell JT. Collagen in abdominal aortic aneurysm: typing, content, and degradation. J Vasc Surg 1987; 6: 578.
35. Koch AE, Haines GK, Rizzo RJ, et al. Human abdominal aortic aneurysms. Immunopathologic analysis suggesting an immune-mediated response. Am J Pathol 1990; 137: 1199.
36. Brophy CM, Reilly JM, Smith GJ, Tilson MD. The role of inflammation in nonspecific abdominal aortic aneurysm disease. Ann Vasc Surg 1991; 5: 229.
37. Baxter BT, McGee GS, Shively VP, et al. Elastin content, cross-links, and mRNA in normal and aneurysmal human aorta. J Vasc Surg 1992; 16: 192.
38. Chambers WH, Vujanovic NL, De Leo AB, Olszowy MW, Herberman RD, Hiserodt JC. Monoclonal antibody to a triggering structure expressed on rat natural killer cells and adherent lymphokine activated killer cells. J Exp Med 1989; 169: 1373.
    39. Dijkstra CD, Dopp EA, Joling P, Kraal G. The heterogeneity of mononuclear phagocytes in lymphoid organs: distinct macrophage subpopulations in the rat recognized by monoclonal antibodies ED1, ED2 and ED3. Immunology 1985; 54: 589.
      40. Bellon B, Capron M, Druet E, et al. Mercuric chloride induced autoimmune disease in Brown-Norways rats: sequential search for antibasement membranes antibodies: Eur J Clin Invest 1982; 12: 127.
        41. Schulze M, Baker PJ, Perkinson DT, et al. Increased urinary excretion of C5b-9 distinguishes passive Heymann nephritis in the rat. Kidney Int 1989; 35: 60.
          © Williams & Wilkins 1996. All Rights Reserved.