3-Deazaadenosine Prevents Leukocyte Invasion by Suppression of Adhesion Molecule Expression During Acute Cardiac Allograft Rejection: Involvement of Apoptotic Cell Death
Horst Fingerhuth, DVM,a Hans Ho¨lschermann, MD,a Helmut Grimm, MD,b Harald Tillmanns, MD,a Werner Haberbosch, MD,a Ruediger C. Braun-Dullaeus, MD,a,c and Thomas H. W. Stadlbauer, MDa
Background: In the initial phase after cardiac transplantation, mononuclear cells infiltrate the graft, initiating a relevant impulse for rejection. 3-Deazaadenosine (c3Ado), an analog of adenosine, has proven anti-inflammatory properties both in vitro and in vivo. We hypothesized that c3Ado can serve as a therapeutic tool to reduce cellular infiltration in cardiac allograft transplantation.
Methods: Using the Wistar–Furth-to-Lewis rat cardiac allograft model, animals were treated with 5 mg c3Ado subcutaneously twice per day. Allografts of untreated animals served as controls. Grafts were harvested on Days 1, 3 and 6 after transplantation for further examination (n = 4 per group and timepoint).
Results: Immunohistochemical examination of c3Ado-treated grafts revealed up to 80% reduction of infiltrating major histocompatability complex (MHC) II–positive cells and T-cell-receptor-positive cells (R73) as well as ED1-positive monocytes and macrophages at Days 3 and 6 after transplantation. Adhesion molecule (ICAM-1 and VCAM-1) expression at Days 1 and 3 was almost completely abolished in c3Ado-treated grafts. However, c3Ado treatment did not prevent apoptotic cell death (TUNEL assay, DNA laddering) at Day 6, nor did it prolong allograft survival. As in controls, grafts were rejected at Day 7.
Conclusion: c3Ado significantly reduces graft infiltration by preventing leukocyte invasion, most likely through suppression of adhesion molecule expression. Although graft survival was not prolonged, treatment with c3Ado may still serve as a strategy to protect hearts from early damage after transplantation. Further studies will show whether peri-operative use of c3Ado can bridge the critical phase after transplantation when standard immunosuppression is not yet completely efficacious.
Solid-organ transplantation has become the treatment of choice for many patients with end-stage organ fail- ure. Despite improvements in immunosuppressive ther- apy, acute rejection continues to occur within the first days or weeks after engraftment, with a frequency of approximately 1.3 episodes per patient during the first year after transplantation,1–3 and these episodes remain a prominent cause of post-transplant morbidity, mortal- ity and graft loss.4 Multicenter studies5 as well as single-center studies6,7 have reported acute rejection episodes in up to 100% of cardiac allograft recipients receiving cyclosporine-based immunosuppressive ther- apy. Acute rejection episodes are known to predispose for development of chronic rejection and chronic allograft dysfunction8 and, therefore, have a tremendous impact on long-term results after solid-organ transplantation.
Clinically, moderate acute rejection episodes develop unnoticeably and without allograft dysfunction, whereas severe acute rejection episodes are character- ized by deterioration of graft function. Mechanistically, acute allograft rejection is primarily T-cell dependent and has been considered a lymphocyte-directed immu- nologic event, the result of an allogeneic immune response of the recipient’s immune system to the graft.9,10Histologically, the most striking feature of acute rejection is progressive infiltration of the graft by mononuclear cells.11,12 These originally circulating host cells of recipient origin interact with graft antigens and activate effector cells that enter the graft via the circu- lation and trigger graft destruction by attracting large numbers of non-specific lymphocytes and macro- phages. For the initial interaction of these different cells, which orchestrate the rejection cascade that leads to graft destruction and ultimate graft loss, adhesion molecules from the selectin family, intercellular adhe- sion molecule-1 (ICAM-1) and vascular cell adhesion molecule-1 (VCAM-1), and cytokines like tumor necro- sis factor-α (TNF-α) and interleukin-1 (IL-1) play a central role.13,14
Our knowledge of the actual events by which host cells destroy a graft remains incomplete. The many proposed effector mechanisms of acute allograft rejec- tion include direct cytotoxicity of primarily CD8+ cells, delayed-type hypersensitivity-induced graft injury, T- cell–mediated graft rejection, natural killer cell activa- tion and alloantibody-mediated graft destruction.15
It has been suggested that effector pathways of graft destruction may act through the induction of apopto- sis,16,17 a programmed form of cell death, at the level of the graft. Therefore, induction of apoptosis seems to play an important role in graft destruction in acute rejection.The relevance of nitric oxide (NO) in the context of acute cardiac allograft rejection is still in the process of investigation.18,22,23 In association with apoptotic cell death, excessive production of NO has been observed during acute cardiac allograft rejection at the level of the graft18,24; in addition, increased serum levels of this mediator have been found during allograft rejection.
3-Deazaadenosine (c3Ado), a structural analog of adenosine, has been shown to prevent reactive oxygen species (ROS) production and has demonstrated anti- inflammatory properties in vitro.27–32 We have previ- ously demonstrated that c3Ado is highly effective in preventing atherosclerotic lesion formation in C57BL/6J mice through inhibition of adhesion molecule expression, with subsequent reduction in monocyte and macrophage recruitment.33 In addition, c3Ado is able to protect lipopolysaccharide (LPS)-induced im- pairment of myocardial function by prevention of nu- clear factor kappa B (NF-nB) activation, inhibition of adhesion molecule expression, and subsequent myocar- dial leukocyte infiltration. This resulted in reduced local production of NO and ROS by accumulated leuko- cytes.34 In addition, c3Ado was able to protect lungs from ischemia and reperfusion injury in combination with University of Wisconsin solution.The aim of our study was to investigate the effects of c3Ado on acute cardiac allograft rejection in a fully allogeneic cardiac allograft model and to further dissect mechanisms of cardiac allograft rejection.
METHODS
Animals
Inbred male Lewis (LEW, RT-1l) and Wistar–Furth (WF, RT-1u) strain rats were purchased from Harlan Winkel- mann (Borchen, Germany) and housed in the animal care facilities at the Department of Surgery, University of Giessen, Giessen, Germany. The animals were kept under standard temperature, humidity and timed light conditions and were given rat chow and water ad libitum. Animals were treated in a humane manner in compliance with the Guide for the Care and Use of Laboratory Animals (NIH Publication No. 86-23, re- vised 1985). The transplant procedures were per- formed when animals attained a body weight of ~250 g. LEW rats served as recipients of WF cardiac allografts.
Cardiac Allograft Model
Hearts were transplanted to the infrarenal great vessels by standard microvascular techniques, as previously described.36,37 Briefly, after each recipient and donor rat was anesthetized by intramuscular injection of 5% ketamine and 2% xylazine, the abdom- inal great vessels of the donors and recipients were carefully separated. The donor heart was harvested and flushed with 4°C cold saline solution. The donor aorta was then anastomosed to the recipient abdom- inal aorta with 7-0 Prolene in an end-to-side manner, followed by anastomosis of the donor pulmonary artery to the recipient inferior vena cava in an end-to-side manner. The tourniquets around the re- cipient great vessels were released and the graft was then allowed to be perfused. Within seconds the grafted hearts began beating. Thereafter, the abdo- men was closed by sutures and graft function was monitored by daily palpation. Rejection was defined as the day of cessation of myocardial contraction. One group of animals was treated with c3Ado (5 mg/kg subcutaneously; Southern Research, Birming- ham, AL) dissolved in 0.9% NaCl every 12 hours. Unmodified LEW recipients of WF cardiac allografts served as the acute rejection control group. For further analysis, grafts from both groups were har- vested at Days 1, 3 and 6 after transplantation and sub-divided.
Immunohistochemical Analysis
Two representative slices of the explanted cardiac allografts were embedded in isopentane-pre-cooled OCT compound (Miles, Elkhart, IN), quick frozen in liquid nitrogen, and stored at —80°C until immunohis- tologic evaluation.
Antibodies
Mouse monoclonal anti-rat ICAM-1 (CD54/Seikagaku, Tokyo, Japan), goat polyclonal anti-human VCAM-1 (Santa Cruz, Santa Cruz, CA), anti-rat ED1, anti-rat MHC II and anti-rat T-cell receptor (TCR)-α/β (Serotec, Ox- ford, UK) and rabbit polyclonal anti-rat inducible nitric oxide synthase (iNOS; Dianova, Hamburg, Germany) were used in this study.
After fixation in ice-cold acetone for 10 minutes, cryostat sections (6 µm) were dried and incubated in a 1:750 dilution of rat serum (Sigma, St. Louis, MO) for 10 minutes. After rinsing with RPMI-1640 (Gibco, Rock- ville, MD), sections were incubated with first antibody (dilution 1:10 for MHC II, 1:20 for VCAM-1, 1:75 for ICAM-1 and TCR-α/β, 1:150 for ED1 and 1:500 for iNOS) for 45 minutes at room temperature. After addi- tional washing steps in Tris buffer, sections were exposed to secondary antibody (rabbit anti-mouse IgG: ICAM-1, ED1, MHC II, TCR-α/β [1:200]; mouse anti-goat IgG: VCAM-1 [1:300]; mouse anti-rabbit IgG: iNOS [1:500]) for 10 minutes, followed by incubation with a linking antibody, as appropriate (rabbit anti-mouse IgG: VCAM-1 [1:400] and iNOS [1:200]) for 10 minutes. An alkaline phosphatase anti-alkaline phosphatase complex (APAAP) (1:50, Dianova, Hamburg, Germany) was added for 20 minutes. Sections were then developed in New Fuchsin solution (30 min) and counterstained with hemalaun (Merck, Darmstadt, Germany) for 30 seconds. Control sections were treated with linking secondary antibody and APAAP complex only.
For quantification, the intensity of ICAM-1 and VCAM-1 staining of 1 section per slice of tissue was scored microscopically in the venules in a blinded fashion (from 0 to 3 as follows: 0 = no staining; 1 = weak staining; 2 = moderate staining; 3 = strong staining).
Both the infiltrating cells and iNOS-positive cells were quantified by counting the antigen-positive cells in 10 fields (magnification ×400) in sections of both slices by blinded, independent investigators.
DNA Laddering
The DNA was extracted from heart tissue specimens with a DNA laddering kit (R&D Systems, Wiesbaden, Germany), according to the manufacturer’s instructions. Briefly, after extraction, 10 µg of DNA was electrophoresed on a 1.5% agarose gel for 3 hours at 80 V. The DNA fragments were visualized with ethidium-bromide staining and illuminated with ultraviolet (UV) light.
TUNEL Assay
DNA strand breaks in situ were detected using the AP In Situ Cell Death Detection Kit (Roche, Molecular Bio- chemicals, Mannheim, Germany). In brief, after fixation in formalin solution and permeabilization in Triton X-100 buffer (0.1% Triton X-100, 0.1% sodium citrate), 6-µm sections were rinsed with labeling solution (TdT- transferase and fluorescein labeled nucleotide mixture in reaction buffer) and incubated for 60 minutes at 37°C. The fluorescein-labeled nuclei were detected under a fluorescence microscope (magnification ×400, 450 to 500 nm).38,39
Statistical Analysis
Statistical analysis was performed using statistical software (STATSDIRECT, version 1.9.8, CamCode, Ashwell, UK). Re- sults were expressed as mean ± SD, unless specified otherwise. To analyze statistical differences between ex- perimental and control groups, the log-rank test and the Mann–Whitney U-test were applied as indicated. p < 0.05 was considered statistically significant. RESULTS Acute Cardiac Allograft Rejection The study was performed in the WF-to-LEW heterotopic rat cardiac transplant model. This is a fully allogeneic strain combination that includes MHC Class I and II mismatches. Both untreated control animals and c3Ado- treated LEW recipients of WF cardiac allografts rejected grafts in an acute fashion within 6 to 8 days, without significant differences in mean survival times (6.0 ± 0.7 days vs 5.3 ± 0.5 days; log-rank test, not significant). For further analysis, functioning WF cardiac allografts with or without c3Ado treatment were removed at Days 1, 3 and 6 (n = 4 for each timepoint and group) after transplantation. Native WF cardiac allografts served as controls (n = 4).Immunohistochemical Analysis of Intragraft Events: ICAM-1, VCAM-1, ED1, MHC Class II, R73 and iNOS. Expression Immunohistochemical analysis of untreated acutely reject- ing cardiac allografts revealed increasing ICAM-1 and VCAM-1 expression at Days 1 and 3 after engraftment (Figures 1 and 2). In addition, untreated grafts showed typical dense infiltration with ED1-positive monocytes and macrophages (Figures 3 and 4) and R73-bearing T cells (Figures 4 and 5), as well as upregulation of MHC Class II molecules (Figure 4) at Days 3 and 6 post-transplantation. Furthermore, notable progressive staining of iNOS-posi- tive cells was observed during the time course of graft rejection (Figure 4). In the c3Ado-treated group signifi- cantly reduced ICAM-1 and VCAM-1 expression in post- capillary venules (p < 0.0001) was observed when com- pared with untreated cardiac allografts at equal timepoints (Figures 1 and 2). Moreover, c3Ado treatment significantly diminished the number of immunocompetent cells infil- trating the graft. Thus, c3Ado prevented invasion of T cells, monocytes and macrophages, as well as iNOS-positive cells, as evidenced by immunocytochemical char- acterization (Figures 3 and 5). Figure 1. Immunolabeling of Day 1 and 3 allografts for ICAM-1 displays significantly diminished expression in the c3Ado group (E, F) in comparison to acute rejecting cardiac allografts (B, C). Native WF hearts (A) and LPS-stimulated (8 hours) hearts (E) were used as negative and positive controls (original magnification ×400). Figure 2. Immunohistochemical analysis of Day 1 and 3 allografts shows significantly decreased expression of adhesion molecules (A) ICAM-1 and (B) VCAM-1 in the c3Ado-treated group. The bars display the score (0 = no staining, 1 = weak staining, 2 = moderate staining, 3 = strong staining) of ICAM-1 and VCAM-1 expression (in percent). Figure 3. Immunohistochemistry for monocytes and macrophages (ED1) in cardiac allografts on Days 1, 3 and 6. Untreated, acutely rejecting allografts are characterized by an increasing number of positively stained cells from Day 1 (B), to day 3 (C) and to day 6 (D). The c3Ado-treated group shows a significantly reduced number of graft-infiltrating mononuclear cells at all timepoints: (F) Day 1; (G) Day 3; and (H) Day 6. Native WF hearts (A) and LPS-stimulated (8 hours) hearts (E) were used as negative and positive controls (original magnification ×400). Determination of Apoptotic Cell Death In both groups (the c3Ado-treated and untreated acutely rejecting allografts) DNA ladders composed of DNA fragments, multiples of 180 to 220 bp and char- acteristic of apoptosis, were observed at Day 6 after engraftment—the final timepoint of graft loss— but not in the native control hearts (Figure 6). Accordingly, an increased number of TUNEL-positive nuclei were detected at Day 6 in both groups. Native control hearts did not express a large number of TUNEL-positive cells (Figure 7). DISCUSSION The major findings of the present study regarding the impact of c3Ado on acute cardiac allograft rejection are: (1) c3Ado suppresses adhesion molecule expression in the early phase after transplantation. (2) c3Ado reduces cellular infiltration with immunocompetent cells and diminishes the inflammatory response at advanced timepoints of graft rejection. (3) c3Ado treatment re- duces particularly the appearance of iNOS-positive cells at Days 3 and 6 after engraftment. (4) Despite the significant reduction in iNOS-positive and infiltrating cells, c3Ado cannot inhibit the induction of apoptosis and prolong allograft survival. Figure 4. Graft-infiltrating cells on Days 1 to 6 in acutely rejecting recipients vs c3Ado treatment. The numbers of positively stained cells for MHC II and R73 (T cells) as well as ED1 (monocytes and macrophages) and iNOS are significantly reduced in c3Ado-treated animals in comparison to acutely rejecting grafts. Positive cells (bars) are expressed as mean ± SEM of positive cells/field of view (c/FV) (original magnification ×400). p-values assessed by Mann–Whitney U-test. Figure 5. Immunohistochemistry for TCR-α/β–positive T cells (R73) in cardiac allografts on Days 1, 3 and 6. Untreated, acutely rejecting allografts are characterized by an increasing number of positively stained cells from Day 1 (B), to Day 3 (C) and to Day 6 (D). The c3Ado-treated group shows a significantly reduced number of graft-infiltrating TCR-α/β–positive T cells at all timepoints: (F) Day 1; (G) Day 3; and (H) Day 6. Native WF hearts (A) and LPS-stimulated (8 hours) hearts (E) were used as negative and positive controls (original magnification ×400). The immunobiology of acute graft rejection and possible therapeutic strategies to prevent graft rejection require further study, as the total number of transplant survivors rises each year with >60,000 cardiac trans- plantations having been performed worldwide and acute rejection continuing to be a major cause of death within the first year after heart transplantation.4 More- over, acute rejection episodes, especially in the early phase after transplantation, represent a significant risk factor for the development of chronic graft vasculopa- thy as a consequence of response to injury.8 Therefore, development of adjunctive treatment strategies to pre- vent acute rejection episodes, especially during the critical phase immediately after transplantation when standard immunosuppressive therapy is not yet com- pletely effective, is of tremendous importance.
In the critical phase initiating acute rejection, adhe- sion molecules like ICAM-1 and VCAM-1 play a central role in mediating cell-to-cell interactions between the graft and immunocompetent cells.13 These adhesion molecules, which are involved at all levels of the immune response and inflammation, play an important role in the initiation of the immunologic response after transplantation.40 Accumulating evidence suggests that inhibition of adhesion molecule expression may be a powerful treatment modality to suppress or ameliorate acute rejection by attenuating attraction of immunocompetent cells. To address this question, the inhibi- tion of adhesion molecule interaction has been investi- gated extensively in animal models. Using antibodies to ICAM-1 and/or LFA-1, reperfusion injuries could be diminished.41 Administration of antibodies targeting adhesion molecules was shown to attenuate cellular and vascular rejection in transplant models.42–45 Hence, in some protocols, prolongation of cardiac allograft survival could be accomplished by anti–LFA-1 monoclo- nal antibody therapy.42 To inhibit adhesion molecule expression in acute cardiac allograft rejection in rats on a transcriptional level, decoys to NF-nB were trans- fected into the allografts, leading to prolonged allograft survival, which reduced neointimal proliferation in a chronic rejection model.
Figure 6. Laddering of DNA fragments that are multiples of 180 to 220 bp is apparent in Day 6 allografts of both groups, but not in the native heart (n = 4).
Figure 7. Representative fluorescence photomicrographs (original magnification ×400; 450 to 500 nm) of heart tissue sections of acutely rejecting allografts vs c3Ado-treated grafts. The number of TUNEL-positive cells is increased in Day 6 allografts of both groups compared with the native hearts.
The prevention of adhesion molecule expression is one of the best-characterized mechanisms of action of c3Ado.27,33,34 In our study, significant suppression of adhesion molecule expression in the c3Ado-treated grafts was demonstrated at Days 1 and 3. In parallel, decreased adhesion molecule expression on the vascu- lar endothelium reduced the adherence of immuno- competent cells (e.g., T cells and macrophages) and their invasion into the graft.
For these immunocompetent cells the main targets of alloimmune response are the MHC antigens. Therefore, the early interaction between donor MHC antigen– peptide complex and T-cell receptors is a critical part of the immune response. The first cells interacting with foreign tissue are T-helper cells, triggering activation of the immunologic response by cytokine and chemokine production,15,47 followed by lymphocyte infiltration and attraction of macrophages, resulting in organ de- struction by different effector mechanisms.
We demonstrated that c3Ado could significantly re- duce the number of infiltrating T cells at Days 3 and 6 after engraftment. In addition, cellular infiltration with macrophages was attenuated on Days 3 and 6. This resulted in diminished production of cytokines and therefore inhibition of graft rejection. However, similar to our findings, Renkonen et al reported that a dramatic reduction in the number of graft-infiltrating cells did not prolong cardiac allograft survival in a fully allogeneic cardiac transplant model. Their approach to block selectins with peptides reduced the histologic hall- marks of acute rejection, but did not affect graft surviv- al.48 This might be due to the fact that reduction of graft-infiltrating cells alone is not sufficient to prevent irreversible graft damage, but the activation level of these cells must also be taken into account. Therefore, induction of additional effector mechanisms ultimately leading to graft loss (e.g., the induction of apoptotic cell death of cardiomyocytes in our rejection model) might play a putative role.
One of the mediators of tissue injury that might be toxic is nitric oxide (NO). NO is a reactive oxygen species that can damage DNA and induce cell death. Enhanced levels of iNOS, increasing serum levels of NO products and elevating enzyme activity, have been documented during allograft rejection.24,26 In- creased NO production was shown to cause direct depression of contractility in an isolated heart mod- el.49 Therefore, the specific reduction of iNOS-posi- tive cells in the graft through c3Ado may have been another possible mechanism of protecting the graft from early injury.
Szaboles et al described the parallelism of iNOS induction and apoptosis and the major role of apoptosis of cardiac myocytes as well as apoptosis of leukocytes and endothelial cells during the course of cardiac allograft rejection.18,23 However, in our experimental fully allogeneic cardiac allograft model, the induction of apoptosis—and thus ultimatively graft loss— could not be prevented by c3Ado, despite the significant reduc- tion of both infiltrating cells and iNOS-positive cells. This is surprising, because induction of apoptosis has been associated with activation of cytotoxic T cells,12,50 as well as with iNOS production of infiltrating macro- phages and cardiac myocytes.18 The Fas-ligand-bearing graft-infiltrating cells have been suspected to be mainly responsible for triggering apoptosis during allograft rejection.21
The failure of c3Ado to prevent induction of apopto- sis and its insufficiency in prolonging cardiac allograft survival support the hypothesis that apoptotic cell death is a major cause of graft destruction. Therefore, further investigations are warranted with regard to the specific pathways inducing apoptotic cell death in the allogeneic transplant setting.
Nevertheless, c3Ado demonstrated a strong effect in preventing expression of adhesion molecules and graft infiltration with immunocompetent cells, a hallmark of acute rejection, despite the fact that c3Ado was not able to prolong cardiac allograft survival. Acute rejection episodes are known to cause structural damage, includ- ing cardiomyocyte loss, and are a significant risk factor for the development of chronic graft vasculopathy as a consequence of response to injury.8 Treatment with c3Ado may serve as a novel strategy to protect grafts from early damage after transplantation and to preserve organ function.
In conclusion, c3Ado has been shown to be a potent agent in suppressing adhesion molecule expression and, consequently, reducing graft infiltration. Specifi- cally, the number of local T cells, macrophages and iNOS-positive cells present during acute cardiac allo- graft rejection could be diminished. c3Ado may serve as a novel strategy to protect grafts from early damage after transplantation and to preserve organ function. Future studies will elucidate whether peri-operative use of c3Ado can help to bridge the critical phase after transplantation when standard immunosuppression is not yet completely effective.