Intimal Thickening Develops Without Humoral Immunity in a Mouse Aortic Allograft Model of Chronic Vascular Rejection
Background The major threat to the long-term survival of cardiac allograft recipients is the development of diffuse intimal thickening in the allograft coronary arteries through mechanisms that are poorly understood. Although antidonor antibodies have been associated with the development of this condition, a causal relationship has not been established.
Methods and Results To determine whether humoral immune responses are necessary for the development of graft vascular disease, we performed abdominal aortic allografts from normal donor mice into different immunodeficient recipient mice: those lacking all donor-specific immune responses (severe combined immunodeficient [SCID] mice and recombination activating gene-1 [RAG-1]–deficient mice) and those lacking humoral immune responses alone owing to a targeted deletion of the joining region (JH) gene segments for the immunoglobulin heavy chain. At 6 to 9 weeks after transplantation, aortic allografts in normal immunocompetent recipients showed concentric intimal thickening extending the full length of the graft (percent luminal reduction, [%LR], 31.2±9.1 [mean±SD] and 38.5±3.6 in different donor-recipient strain combinations). In contrast, syngeneic (histocompatible) aortic grafts showed a normal-appearing vessel wall (%LR, 1.6±0.7). In both SCID and RAG-1–deficient recipients, aortic allografts showed a virtual absence of neointimal formation (%LR, 3.7±2.1 and 3.8±1.6 in SCID and RAG-1–deficient recipients, respectively), indicating a critical etiological role for alloimmune responses in this model. Importantly, allografts in JH-deficient mice showed marked intimal thickening (%LR, 35.7±7.9), with an appearance histologically indistinguishable from that of normal immunocompetent recipients.
Conclusions Neointimal formation in graft vascular disease is critically dependent on alloimmune responses of the host. Humoral effector mechanisms, however, may not be required.
The origin of diffuse obliterative intimal thickening that develops in the blood vessels of transplanted human hearts remains inadequately understood. This pathological process persists despite conventional immunosuppression and is the leading cause of late graft failure in human heart transplantation. A seminal study by Hess et al1 showed that the presence of cytotoxic anti–HLA-DR antibodies was associated with an increased incidence and severity of allograft coronary disease. Hammond et al2 later reported that an accumulation of immunoglobulin and complement in and about the coronary microvasculature was associated with a 10-fold increase in the 5-year risk of angiographic coronary disease. These studies were important in establishing a strong association with the humoral immune response of the recipient but were limited by their observational nature. Furthermore, Hosenpud et al3 in a recent study found that allograft coronary disease correlated much better with cell-mediated reactivity against donor vascular endothelium than with humoral immunity in the form of donor-specific anti-endothelial antibodies.
To investigate the role of humoral immune responses in the development of allograft intimal thickening, we studied abdominal aortic allografts in recipient mice with or without specific immunodeficiencies. We found that significant intimal thickening occurred in mouse aortic allografts in the absence of humoral immune effector mechanisms.
All mice were used in accordance with the guidelines of the Council on Animal Care of the University of Western Ontario (Canada). Normal adult male and female mice of BALB/c (H-2d), C57BL/6 (H-2b), and 129/Sv (H-2b) strains (Jackson Laboratory) were used, along with homozygous immunodeficient mice of three different genotypes as follows: (1) C3H/HeSn (H-2k) strain mice with severe combined immunodeficiency (SCID; bred from trios; Jackson Laboratory), (2) C57BL/6 strain mice with recombination activating gene (RAG)-1 deficiency (C57BL/6J-rag1m1Mom; Jackson Laboratory), and (3) 129/Sv strain mice with a targeted deletion of the joining region (JH) genes of the immunoglobulin heavy chain (gift of GenPharm International). SCID mice have a spontaneous defect in V(D)J recombinase activity that prevents productive immunoglobulin and T-cell–receptor gene rearrangement. RAG-1–deficient mice, derived by targeted gene disruption in embryonic stem cells, are phenotypically identical to SCID mice but are more stable in their blockade of lymphocyte differentiation. Moreover, in contrast to the C3H/HeSn SCID mice, which were smaller, the adult weight of C57BL/6 RAG-1–deficient recipients consistently exceeded a threshold of 22 to 25 g required for successful aortic graft surgery. JH-deficient mice manifest a selective loss of all B-cell function because of a failure to complete the limiting first step in immunoglobulin gene rearrangement.4 5 The phenotypic integrity of the immunodeficient mice was confirmed by a total serum immunoglobulin concentration <0.01 g/L, an absence of surface-IgM–bearing splenocytes in immunostained sections, and hypoplastic appearance of the lymphoid organs, including an absent thymic cortex in SCID and RAG-1–deficient mice.
Aortic Graft Surgery and Graft Retrieval
Donor and recipient mice were anesthetized with 45 mg/kg IP pentobarbital and 0.1 mg/kg SC buprenorphine. A 3-mm segment of donor infrarenal aorta was grafted into the recipient infrarenal aorta by end-to-end anastomosis. No immunosuppressive medication was used. Aortic grafts were removed under anesthesia 6 to 9 weeks later, after fixation by perfusion at mean physiological pressure with 4% paraformaldehyde administered by catheter through the thoracic aorta. The perfusion-fixed graft was paraffin-embedded and sectioned 5 μm thick for standard histological staining. At least eight transverse sections spread out along each graft were obtained for study. In some cases, the grafts were perfused with saline instead of fixative and were flash-frozen in OCT medium (Tissue-Tek) in liquid nitrogen for immunohistochemical staining in cryostat sections.
Determination of Percent Luminal Reduction
Luminal reduction was assessed in perfusion-fixed sections by use of computer-assisted video microscopy (Jandel Scientific). Percent luminal reduction was defined as [(area bounded by the internal elastic lamina minus area of the lumen)/(area bounded by the internal elastic lamina)]. The coefficient of variation was 3%. The percent luminal reduction for each graft was averaged from the measurement of three to four representative sections, with all measurements made with investigators blinded to recipient group identity.
A standard avidin-biotin immunoperoxidase method was performed with an Elite Vectastain kit (Vector). Primary antibodies consisted of monoclonal rat anti-mouse antibodies directed against the following inflammatory cell markers: Thy-1 (all T cells; clone 30H12; Becton-Dickinson); CD4 (MHC-class-II–restricted T cells; clone GK1.5; Becton-Dickinson); CD8 (MHC-class-I–restricted T cells; clone 53-6.7; Becton-Dickinson); Mac-1 (macrophages; clone M1/70; Boehringer-Mannheim); and immunoglobulin heavy-chain μ (IgM-bearing B cells; clone Ig8; Cedarlane). In addition, monoclonal antibody to a synthetic decapeptide of smooth muscle α–actin (clone 1A4; BioGenex) was used to identify vascular smooth muscle cells (SMCs). Negative controls were performed by substituting blocking solution for the primary antibody. Naive spleen and thymus sections of immunocompetent mice served as positive controls for inflammatory cells; normal aortic media similarly controlled for vascular SMCs.
Numerical data are expressed as mean±SD. Data on percent luminal reduction were analyzed by nonparametric ANOVA (Kruskal-Wallis), with post hoc assessment of individual differences by Bonferroni-adjusted Mann-Whitney U test. A value of P<.05 (two-tailed) was considered statistically significant.
The Figure⇓ provides the major findings in the different experimental groups; the Table⇓ provides details on donor-recipient strain combinations, numbers of grafts, and luminal reduction. All grafts were obtained from normal donor mice with the following groups of recipient mice: (1) syngeneic (histocompatible; no rejection), (2) allogeneic (histoincompatible) but otherwise normal and immunocompetent (control allografts), (3) allogeneic and deficient for both T- and B-cell function (SCID mice and RAG-1–deficient mice), and (4) allogeneic and deficient for B-cell function alone (JH-deficient mice).
Syngeneic BALB/c-to-BALB/c aortic grafts developed no graft vascular disease (the Figure,⇑ part a), maintaining a healthy appearance and a normal intimal layer (high magnification not shown). Importantly, however, BALB/c aortic segments were fully susceptible to neointimal formation when grafted into allogeneic recipients (C57BL/6; see below), and the BALB/c host was fully supportive of neointimal formation when grafted with allogeneic aortic segments (C57BL/6; data not shown). Thus, operative manipulation alone was insufficient to cause graft vascular disease.
Allogeneic aortic grafts showed well-established intimal thickening by 6 weeks after transplantation (BALB/c-to-C57BL/6; the Figure,⇑ part b). Histological features included a circumferential neointima that involved the full length of the graft; a highly cellular composition, including especially SMCs; and an extracellular matrix rich in proteoglycans (the Figure,⇑ parts b through d). Importantly, these are also characteristic of human allograft coronary disease. At 6 to 9 weeks after transplantation, the aortic allografts had a pleomorphic inflammatory cell response that was concentrated in the adventitia, with sparse mononuclear inflammatory cells infiltrating the neointima and consisting of approximately equal numbers of macrophages and CD4+ and CD8+ T cells. The aortic allograft media was severely atrophic, and the elastin layers appeared damaged. In the mice, as in human patients, the graft vascular disease stopped at the anastomoses, sparing the host vessels.
A major pathogenic role of host alloimmune response was demonstrated in immunodeficient aortic allograft recipients that lacked both T- and B-cell functions (SCID and RAG-1–deficient mice). In these recipients at 6 to 9 weeks after transplantation, all layers of the allograft vessel wall resembled syngeneic grafts except for minute regions of intimal thickening, often appearing as no more than isolated vascular SMCs located abluminally (the Figure,⇑ part e).
In marked contrast, in allograft recipients lacking humoral immunity alone (JH-deficient mice), severe graft vascular disease was apparent by 6 weeks after transplantation (the Figure,⇑ part f). This response was comparable in nature and severity to that of normal recipient mice of the same strain background (129/Sv; the Table⇑).
We have shown that host alloimmune function plays a major pathogenic role in the development of graft vascular disease in mouse aortic allografts. Furthermore, in this model, antibody-mediated immunity is not required in the development of graft vascular disease, contrary to the concept of graft vascular disease as a form of chronic humoral rejection. Others have shown that blockade of tumor necrosis factor-α reduces T-cell infiltration and neointimal formation in rabbit cardiac allografts.6 Similarly, blockade of the CD28-B7 costimulatory pathway of T-cell activation modulates T-cell and macrophage activation and attenuates intimal thickening in rat heart allografts.7 Together with our data, these studies support a pivotal role for cellular immunity in graft vascular disease.
In contrast, a report by Russell et al8 using mouse heterotopic heart grafts has emphasized humoral rejection. They showed that allograft coronary disease occurred in one donor-recipient combination but not in the reverse combination in which the recipients could not generate cytotoxic antibodies to donor major histocompatibility complex class I antigens. The paradigm is complex, however, because these investigators have reported other data in mouse heterotopic heart grafts highlighting T-cell–dependent mechanisms.9 Importantly, aortic grafts differ significantly from heterotopic heart grafts, especially for graft vascular disease. First, the transplanted aortic segments are functional and fully load-bearing, unlike the coronary arteries of intra-abdominal heart grafts that do not support the circulation. Second, the aortic grafts are unaffected by growth factors and vasoactive molecules from thrombi that frequently occupy the sequestered left ventricle. Finally, the aortic grafts are unaffected by the inflammatory cascade inherent in acute cellular rejection of the myocardium, which may or may not be relevant to the development of graft coronary disease in humans. Thus, the models are distinct, although the summed data do suggest that the environment may modify the pathogenesis involved.
Shi et al10 recently devised a mouse model of carotid loop grafts, performed by end-to-side anastomosis, with the native carotid left untied. With this model, they reported that multiple immunological pathways are important in graft vascular disease, including B cells, CD4+ T cells, and macrophages. Their results thus differ from ours with respect to the role of humoral immunity. Carotid loop grafts, however, differ significantly from aortic grafts. Without end-to-end anastomosis, the carotid loops create a flow divider with altered shear forces that affect many elements of neointimal formation such as platelet deposition, inflammatory cell adhesion, and endothelial cell dysfunction. Moreover, as Shi et al10 suggested, SMC responsiveness was likely the ultimate determinant of the extent of intimal thickening in the carotid loops; vascular SMCs, in turn, do not represent a single population but have notable phenotypic heterogeneity that depends on many factors, including their location in the arterial tree.11 These distinctions may account for the partial differences observed between carotid loops and aortic grafts.
In summary, our findings clearly support host alloimmune responses as the dominant initiating factor in graft vascular disease, but humoral immunity may not be essential in its pathogenesis.
This study was supported by a grant from the Heart and Stroke Foundation of Ontario, a Career Scientist Award of the Ontario Ministry of Health (Dr Chow), and a Research Scholarship of the Medical Research Council of Canada (Dr Pickering). We thank K. James and M.A. Laurin for skilled technical assistance.
- Received August 19, 1996.
- Revision received October 7, 1996.
- Accepted October 9, 1996.
- Copyright © 1996 by American Heart Association
Hess ML, Hastillo A, Mohanakumar T, Cowley MJ, Vetrovac G, Szentpetery S, Wolfgang TC, Lower RR. Accelerated atherosclerosis in cardiac transplantation: role of cytotoxic B-cell antibodies and hyperlipidemia. Circulation. 1983;68(suppl II):II-94-II-101.
Hosenpud JD, Everett JP, Morris TE, Mauck KA, Shipley GD, Wagner CR. Cardiac allograft vasculopathy: association with cell-mediated but not humoral alloimmunity to donor-specific vascular endothelium. Circulation.. 1995;92:205-211.
Chen J, Trounstine M, Alt FW, Young F, Kurahara C, Loring JF, Huszar D. Immunoglobulin gene rearrangement in B cell deficient mice generated by targeted deletion of the JH locus. Int Immunol.. 1993;5:647-656.
Jakobovits A, Vergara GJ, Kennedy JL, Hales JF, McGuinness RP, Casentini-Borocz DE, Brenner DG, Otten GR. Analysis of homozygous mutant chimeric mice: deletion of the immunoglobulin heavy-chain joining region blocks B-cell development and antibody production. Proc Natl Acad Sci U S A.. 1993;90:2551-2555.
Clausell N, Molossi S, Sett S, Rabinovitch M. In vivo blockade of tumor necrosis factor-α in cholesterol-fed rabbits after cardiac transplant inhibits acute coronary artery neointimal formation. Circulation.. 1994;89:2768-2779.
Russell ME, Hancock WW, Akalin E, Wallace AF, Glysing-Jensen T, Willett TA, Sayegh MH. Chronic cardiac rejection in the LEW to F344 rat model: blockade of CD28-B7 costimulation by CTLA4Ig modulates T cell and macrophage activation and attenuates arteriosclerosis. J Clin Invest.. 1996;97:833-838.
Russell PS, Chase CM, Winn HJ, Colvin RB. Coronary atherosclerosis in transplanted mouse hearts. II: importance of humoral immunity. J Immunol.. 1994;152:5135-5141.
Shi C, Lee W-S, He Q, Zhang D, Fletcher DL Jr, Newell JB, Haber E. Immunologic basis of transplant-associated arteriosclerosis. Proc Natl Acad Sci U S A.. 1996;93:4051-4056.
Gadson PF Jr, Rossignol C, McCoy J, Rosenquist TH. Expression of elastin, smooth muscle alpha-actin, and c-jun as a function of the embryonic lineage of vascular smooth muscle cells. In Vitro Cell Dev Biol Animal. 1993;29A:773-781.