*Abbreviations: BSA, bovine serum albumin; CsA, cyclosporine; hDAF, human decay-accelerating factor; HAR, hyperacute rejection; H&E, hematoxylin and eosin; IVC, inferior vena cava; mAb, monoclonal antibody; PBS, phosphate-buffered saline; PNAb, preformed natural antibodies; RBC, red blood cell; RCA, regulator of complement activation; WBC, white blood cell count.
Xenotransplantation between widely disparate species, such as pig-to-primate, results in hyperacute rejection (HAR*) due to the binding of preformed natural antibodies (PNAb) to α-galactosyl carbohydrate epitopes on vascular endothelial cells, followed by the activation of the complement cascade (1). Alternative pathway activation of complement has also been shown to occur in the pig-to-human xenocombination(2,3). The complement system by itself seems capable of discriminating self from non-self in xenogeneic systems(4,5), which suggests that the blockade of complement activation is a clinically viable strategy of combating HAR.
Regulators of complement activation (RCAs) are species specific(6). Decay-accelerating factor, one of these RCAs, inhibits both the classical and the alternative complement pathway. The production of transgenic pigs expressing human decay-accelerating factor(hDAF) on endothelium has been proposed as a method of blocking HAR(7,8). Transgenic pigs expressing hDAF have been produced, and both ex vivo (9,10) and in vivo(11) data have demonstrated that HAR could be prevented. A heterotopic transgenic pig heart survived for up to 9 weeks in a nonhuman primate (11). These transgenic pig hearts were transplanted into the abdomen of cynomolgus monkeys and, therefore, were not responsible for maintaining the circulation of the recipient.
The aim of the study reported here was to determine whether a pig heart, expressing the hDAF transgene, would sustain cardiac output of a nonhuman primate recipient after an orthotopic, i.e., life-supporting, transplant. To minimize the technical failure rate, the pig-to-baboon model was chosen in preference to the cynomolgus monkey, due to the larger size of the baboons. Previous experiments had demonstrated that hDAF inhibits both cynomolgus monkey and baboon complement, although less efficiently than it inhibits human complement (12).
MATERIALS AND METHODS
Animals and anesthesia. All animals received humane care in compliance with the Code of Practice for Scientific Procedures in Animals formulated by the University of Cambridge and the Guidance on the Operation of Animals (Scientific Procedures) Act 1986 prepared by the Home Office, UK.
Ten heterozygous Large white/Landrace cross-pigs of either sex (age 10-21 days, weight 4.6-8.0 kg) expressing hDAF were used as donors for all experiments. All animals were derived from the A74 founder line, which had been made transgenic for hDAF by micro-injection of a minigene construct (6.5 kb) containing a 4-kb human genomic DNA fragment that incorporates the DAF promoter, the 5′-untranslated and signal peptide sequence of the DAF gene, and the first exon and intron of the gene. This genomic DNA fragment was linked to the cDNA fragment that codes for the remaining exons of the DAF gene (13). hDAF expression in the hearts of A74 offspring was found to be comparable with that seen on the equivalent human tissue (14).
Transgenic animals for this study were identified by DNA slot blot analysis using a random-labeled 32P hDAF cDNA probe. Expression of the transgene was determined by immunohistochemistry as described previously(15).
Donor procedure. After sedation with ketamine hydrochloride (20 mg/kg) and oxygen mask induction, a tracheostomy was performed and mechanical ventilation was begun; general anesthesia was maintained with N2O/O2 (1/1) and Isoflurane (1-2%).
A median sternotomy was performed, the pericardium was opened up longitudinally, and heparin (300 IU/kg) was given intravenously. The inferior vena cava (IVC) was tied above the diaphragm, and incisions were made into the IVC and the right upper lung vein to decompress the heart. The aorta was then cross-clamped at the level of the innominate artery, and 200 ml of St. Thomas cardioplegic solution (4°C) was infused into the aortic root. The heart was excised under topical cooling with saline (4°C) after division of the superior vena cava and IVC, the pulmonary veins, and the ascending aorta. Simple defects such as a patent foramen ovale were repaired.
Recipient procedure. An orthotopic heart transplantation according to the atrial cuff technique first described by Lower and Shumway was performed (16). Ten recipient baboons (Papio anubis, age 2-4 years, weight 6.8-9.3 kg) were sedated using intramuscular ketamine hydrochloride (15 mg/kg) and atropine (0.05 mg/kg) and intravenous diazepam (1 mg/kg). After endotracheal intubation, they were maintained on N2O/O2 (1/1) with the addition of Isoflurane (1-2%) and propofol(50 mg/hr). Analgesia was achieved by intravenous fentanyl citrate. For the priming of the pediatric cardiopulmonary bypass circuit, autologous blood was used that had been obtained 1 and 2 months before surgery. Two units of packed red blood cells (RBC) were separated, frozen at -35°C, and stored at the National Blood Service (Birmingham Centre, UK) until they were defrosted on the day of transplantation. A median sternotomy was used for cardiac excision and graft placement. After implantation and discontinuation of extracorporeal circulation (cardiopulmonary bypass time 94±50 min, ischemic time 62±12 min; Table 1), the chest was closed, leaving a single drainage catheter in the pericardium. If significant bleeding did not occur within 1 hr, the drain was removed and the animal was extubated within 2 to 3 hr after the operation, otherwise the chest was reopened immediately and the source of bleeding was identified. Having completed surgery, the animal was kept in an incubator until it woke up and was subsequently transferred to an oxygen-enriched cage without intravenous fluid replacement or inotropic support.
Immunosuppression. All animals received cyclosporine (CsA) intramuscularly (25 mg/kg on the day of surgery) or orally (starting with 200 mg/kg on the day after transplantation) to achieve trough levels in excess of 600 ng/ml. This CsA target level represents a compromise between experience obtained in clinical allografting and in vitro investigations, which suggest aiming for CsA levels of about 1500 ng/ml to gain comparable effects on baboon lymphocytes compared with human lymphocytes(17).
Cyclophosphamide was given at 40 mg/kg i.v. on the preoperative day, continued by 20 mg/kg i.v. on the day of surgery and thereafter, according to total white blood cell count (WBC; lower limit 2×109/L) and hemolytic anti-pig RBC antibody titers (see below). This minimum WBC count reflects experience that has recently been gained in clinical immunosuppression of patients suffering from autoimmune diseases, such as systemic lupus erythematosus (18). Some authors even recommend a WBC count as low as 1.5×109/L as the minimum acceptable nadir for immunosuppression in patients with lupus nephritis(19).
Prednisolone was started on the day of surgery at 1 mg/kg/day, reducing to 0.2 mg/kg/day by day 18. Rejection episodes were treated with a 3-day course of methylprednisolone at 15 mg/kg/day i.v.
Postoperative monitoring. Blood was sampled daily for hematological examination (Bayer-Technicon H1E haematology analyser), electrolytes (Hitachi 737 clinical chemistry analyser), and trough CsA levels(thin-layer chromatography, Department of Biochemistry, Papworth Hospital). Rejection of the graft was diagnosed by a rise in hemolytic anti-pig-RBC antibody levels and/or by the clinical signs of heart failure (weight increase, edema, dyspnea, cardiac arrhythmia).
Xenoantibodies. Measurement of hemolytic anti-pig RBC antibody: The levels of anti-pig RBC hemolytic antibodies were measured as a means to evaluate both pretransplant and xenograft-induced baboon anti-pig humoral immune responses. In this assay, a pool of four normal human sera (provided by the local Blood Bank, aliquoted, and stored at -70°C) was utilized in each assay as a constant positive control.
Baboon clotted blood samples were spun for 5 min at 3000 rpm at 4°C. Sera were then collected, transferred into Eppendorf tubes, and heat inactivated at 56°C for 30 min. Subsequently, each serum was serially diluted on a 96-well plate with complement fixation diluent (ICN, Costa Mesa, CA) from 1/10 to 1/1280. After the addition of rabbit complement (Sera-Lab, Crawley, UK) and porcine RBC, the plates were placed in an orbital incubator at 37°C for 1 hr. The plates were then centrifuged at 1800 rpm for 10 min, and 100 µl of supernatant from each well were transferred into reading plates. The reading was performed by a Multiskan Plate Reader (ICN) at a wavelength of 420 nm. The mean absorbance was calculated for each sample and expressed in area under curve units, considering the number of area under curve units of the human control as equivalent to 1000.
Measurement of pretransplant anti-α1,3-Gal antibodies(ELISA): Baboon clotted blood samples were spun for 5 min at 3000 rpm at 4°C and stored at -70°C. Anti-α1,3-Gal antibodies were determined using a two-stage assay. Fifty microliters of pretransplant sera were fractionated on a size exclusion column (Waters Protein-PAK 300SW; Waters, Milford, MA) at 1 ml/min in 0.1 M K2HPO4/0.02% sodium azide. The OD 280 was constantly monitored in the effluent, and fractions encompassing the IgM and IgG peaks were collected. Subsequently the IgG and IgM were deposited in triplicate in wells of a microtiter plate coated with HSA-linear B trisaccharide conjugate (Dexta Labs, Reading, UK; 5 µg/ml in carbonate buffer pH 9.6 and blocked with 4% bovine serum albumin [BSA]). A standard series of affinity-purified anti-α1,3-Gal IgG and IgM (diluted in phosphate-buffered saline[PBS]/1% BSA) was deposited on the same plate. The antibodies were allowed to bind for 2 hr at room temperature. Then, the wells were washed with PBS/0.1% BSA and incubated with anti-human whole IgG or IgM-peroxidase conjugate(1:200 in PBS/0.1% BSA; Sigma, St. Louis, MO) for 1 hr at room temperature. The plate was developed with OPD (Sigma) and quenched with 1 M H2SO4, and the absorbance was read at 492 nm. The concentration of anti-α1,3-Gal IgG or IgM in the original serum was calculated by reference to the absorbance given by the standard anti-α1,3-Gal antibodies.
Histopathological examination. After termination of the experiments, all grafts were preserved in formal saline for conventional histological examination using hematoxylin and eosin (H&E) staining. Cardiac tissue was also snap-frozen in liquid nitrogen and stained for deposition of C3, C4, C5b-9 (mouse anti-human monoclonal antibody [mAb]; Dako Ltd., High Wycombe, UK), IgG (mouse anti-human mAb [Dako]), and IgM(mouse anti-human mAb; Immunotech SA, Marseille, France) with the avidin-biotin complex method. Each component was graded in terms of the intensity and the extent of positivity within the specimen (1+ to 4+). Monocytes/macrophages were stained by a mouse anti-human CD68 mAb(Dako); T cells were detected using a mouse anti-human CD4 and anti-human CD8 mAb (Dako), respectively.
Statistics. Statistical analysis was performed using the two-tailed t test for independent samples to compare subgroups(SPSS 7.5. for Windows; SPSS Inc., Chicago, IL). Equal variances were not assumed. Results are reported as mean ± SD. A P value of less than 0.05 was considered a significant difference.
HAR was not observed in any of the 10 transplanted transgenic hearts. Five baboons survived for less than 18 hr; three of them died shortly after graft reperfusion (at 6, 6, and 9 hr) due to failure to produce even a low cardiac output and/or dysrhythmia (Table 1). When these there animals were compared with the other seven baboons, no differences were detected in ischemic time, cardiopulmonary bypass time, and preoperative anti-pig RBC hemolytic antibodies or anti-α1,3-Gal IgM levels. However, pretransplant anti-α1,3-Gal IgG levels were significantly elevated(21.1±6.0 vs. 5.0±4.1 µg/ml; P<0.05). Immunohistological examination revealed no significant increase in the amount of IgG and IgM deposition on endothelium or within interstitial spaces of the myocardium in comparison to the other seven grafts (Table 2). No histological evidence of rejection was observed.
Two animals died of pulmonary artery thrombosis at 10 and 18 hr after reperfusion. No differences in ischemia, cardiopulmonary bypass time, or antibody levels were observed when compared with the other animals, but the donor heart weight exceeded the recipient heart weight by 78.1±28.6% in these two animals. This was in contrast to the other eight animals, in which size mismatch was only 27.7±16.4%. Due to the small group size, the difference was not statistically significant. Again, no histological evidence of rejection was observed in these xenografts.
The causes of death and survival times of the other five animals were as follows: one animal collapsed and was killed because of bronchopneumonia on day 4; three animals rejected their grafts on day 5; and one animal had to be killed on day 9 with a beating xengraft, due to pancytopenia related to the immunosuppressive medication (mean cyclophosphamide dosage 11.2±2.5 mg/kg/day; range 8-14.5 mg/kg/day). In the animal that died of infection, the lowest WBC and lymphocyte count were detected in the early postoperative period (X215; Fig. 1). Two of the three animals rejecting the grafts on day 5 after transplant showed a prominent increase of anti-pig-antibody titers at the time of rejection(Fig. 2).
The two baboons that were killed because of infection or pancytopenia did not demonstrate histological evidence of rejection (Fig. 3). In two of the three rejected grafts, histological examination showed rupture of the endothelial cell layer, edema, thrombosis, hemorrhage, and focal necrosis of the cardiomyocytes. This was compatible with acute vascular rejection. A cellular infiltrate predominantly consisting of mononuclear cells and polymorphonuclear cells was also detected (Fig. 4). In the third graft (X221), a normal myocyte pattern without necrosis and cellular infiltrate but with focal thrombosis of small vessels (Fig. 5) and patchy staining of the terminal complement pathway component C9 was observed (Fig. 6). In all grafts, staining for IgG(Fig. 7) and IgM was positive and the expression of hDAF was preserved, even in the rejected hearts (Fig. 8).
Long-term survival of orthotopic pig-to-primate heart transplants has not yet been reported. Without adsorption of PNAb, normal pig hearts are subject to HAR within 18 hr when transplanted orthotopically into baboons(20).
Recently, progress in genetic engineering techniques has allowed the production of pigs transgenic for hDAF, thus preventing HAR of porcine organs when transplanted into primate recipients (12). Rejection of transgenic xenografts is delayed and seems to be accompanied by induced antixenograft antibodies (21). A combination of CsA, cyclophosphamide, and steroids inhibits the induced antixenograft antibody response in hamster-to-rat transplants (22), and this regimen was therefore used as the first immunosuppressive regimen in transgenic pig-to-primate xenotransplants.
Three animals died of graft failure due to dysrhythmia and/or low output and subsequent systemic acidosis within 9 hr. The limited postoperative care of the recipients without invasive blood pressure monitoring or electrocardiogram makes it difficult to determine the exact cause of early graft failure. Postoperative treatment of the animals did not include inotropic support, therefore, prolonged hypotension might have accounted for the adverse outcome. In orthotopic transplantation of pig cardiac allografts and isografts, Calne et al. (23) achieved a minimum survival of 3 days, with the longest surviving animal living for 1.5 years, proving the technical feasibility of the use of pig hearts as donor organs. No differences in titers of hemolytic anti-pig RBC antibodies or anti-α1,3-Gal IgM antibodies were documented in recipient animals before surgery, however, anti-α1,3-Gal IgG antibodies were significantly elevated. Histological examination showed staining for total IgG both on endothelial cells and within interstitial spaces (Fig. 7), but antibody deposition on the endothelium was similar in all animals, and quantitative differences of antibody staining in interstitial spaces of the myocardium were not demonstrated when these three grafts were compared with the other seven grafts. However, it has been shown in vitro that challenging beating rat cardiomyocytes with PNAb in the absence of complement leads to a temporary standstill (24), and after recovery, to desynchronization for at least 12 hr(25). It can thus be speculated that immunoglobulins exudated into interstitial spaces due to ischemia/reperfusion injury may cover cardiomyocyte cell membranes with an antibody layer. This could impair transmembrane ion fluxes leading to electromyocardial abnormalities(26), especially if binding of antibodies with a higher affinity occurs. This phenomenon would also explain the more favorable outcome of the heterotopic cardiac xenotransplants (21), in which temporary cardiac dysfunction does not lead to a fatal outcome as in the orthotopic transplants. These data emphasize the importance of selecting the correct animal model before considering clinical trials. Previous data from closely related species has demonstrated that prolonged survival of the heterotopic heart is not necessarily confirmed in the orthotopic model(27).
Two hearts failed for technical reasons. Postmortem evaluation revealed pulmonary artery thrombosis at the site of the anastomosis. Size matching in these two cases seemed to be inadequate. The weight of the donor hearts exceeded the native recipient hearts by 58% and 98%, respectively, although the body weights of the donor pigs were 3.5% and 7.8% less than the corresponding recipient baboons. Experience obtained in this study indicates that the heart/body weight ratio of pigs is significantly increased if compared with baboons (0.65±0.08% vs. 0.4±0.04% of the body weight; P<0.001). To match pig and baboon heart sizes, the body weight of donor piglets, therefore, needs to be 20-30% less than the recipient baboon weight.
In the remaining five experiments, two animals received doses of cyclophosphamide that led to serious side effects. Infection occurred in one animal, and the longest survivor was killed on day 9 due to pancytopenia. Rejection was, however, not observed in these two animals. The cyclophosphamide regimen used in this study was based on previous experience obtained in cynomolgus monkeys (11), but seemed to be more toxic to the bone marrow of baboons. Future research will focus on new immunosuppressive drugs to replace or at least reduce cyclophosphamide administration.
Three animals rejected grafts on day 5, possibly as a result of insufficient immunosuppression. Histological findings showed a pattern that was first described in a discordant rodent model of complement-depleted recipients and referred to as acute vascular rejection(28) or delayed xenograft rejection(29). A cellular infiltrate consisting predominantly of monocytes/macrophages (CD68+) and neutrophils was found. CD4+ and CD8+ cells were also present at lower levels. Although marked C4 deposition was detected in all transplanted grafts, C5b-9 was only observed in these three animals, which suggests that terminal complement pathway activation contributed to graft rejection. The data do not directly address this issue, but it is possible that insufficient immunosuppression allowed the formation of high affinity-induced antixenograft antibodies, overwhelming the complement inhibiting properties of hDAF. Although the measurement of hemolytic anti-pig-antibodies in the serum does not accurately reflect what is happening at the level of the xenograft, the antibody spikes that were observed in two of the animals at the time of rejection would support this. The following activation of terminal complement components via the classical pathway seems to play a pivotal role in the pathogenesis of acute vascular rejection early after transplant.
The addition of further human RCAs like CD59 might be able to control this process, in particular if the rejection pattern seems to be focal. In this respect, a human CD59 and hDAF transgenic pig has been produced, and hearts of these double transgenic pigs were transplanted heterotopically into nonimmunosuppressed baboons (30). HAR was not observed and survival times of 6 and 69 hr were recorded. However, organ-specific expression was less than that observed in human tissue, therefore, better results may be obtained in the future with improved expression of these two transgenes.
To achieve prolonged graft survival, the additional use of immunoabsorbent columns (31) for the removal of both preformed naturally occurring antibodies and elicited antigraft antibodies may warrant further investigation. The benefit of PNAb adsorption has previously been shown by Fukushima et al. (20) who extended graft survival of unmodified pig hearts transplanted orthotopically into baboons to a maximum of 16 days. Intraoperatively, PNAb were reduced by hemoperfusion of a pig lung during circulatory arrest. In addition, alternative pathway complement activation was inhibited by the infusion of a complement inhibitor (nafamstat mesylate) at the time of reperfusion (20). The animal eventually succumbed to cytomegalovirus infection.
In a model of heterotopic pig-to-baboon cardiac transplantation, Cooper et al. (32) achieved a maximum survival of 5 days by using preoperative hemoperfusion of donor kidneys before transplantation, while Roslin et al. (33) eliminated PNAb via preoperative perfusion of the donor spleen combined with the use of total lymphoid irradiation. Their longest survivor rejected the graft on day 15 after transplant. It was concluded that graft survival correlated with the effectiveness of the antibody adsorption procedure.
These data also suggest that an immunosuppressive regimen that could safely inhibit the production of induced antixenograft antibodies should result in long-term survival in the hDAF transgenic pig-to-primate model. If successful orthotopic pig-to-nonhuman primate heart transplants become possible, clinical xenotransplants could become a reality.
Acknowledgments. The helpful discussion and comments of Johan van den Bogaerde are gratefully appreciated.
1. Lawson JH, Platt JL. Molecular barriers to xenotransplantation. Transplantation 1996; 62: 303.
2. Zhao Z, Termignon JL, Cardoso J, et al. Hyperacute xenograft rejection in the swine-to-human donor-recipient combination. Transplantation 1994; 57: 245.
3. Makowka L, Cramer DV, Hoffman A, et al. The use of a pig liver xenograft for temporary support of a patient with fulminant hepatic failure. Transplantation 1995; 59: 1654.
4. Atkinson JP, Farries T. Separation of self from non-self in the complement system. Immunol Today 1987; 8: 212.
5. Dalmasso AP, Vercelloti GM, Platt JL, Bach FH. Inhibition of complement-mediated cytotoxicity by decay accelerating factor: potential for prevention of xenograft hyperacute rejection. Transplantation 1991; 52: 530.
6. Atkinson JP, Oglesby TJ, White D, Adams EA, Liszewski MK. Separation of self from non-self in the complement system: a role for membrane cofactor protein and decay accelerating factor. Clin Exp Immunol 1991; 86 (suppl 1): 27.
7. Wang MW, Wright LJ, Sims MJ, White DJG. Presence of human chromosome 1 with expression of human decay-accelerating factor (DAF) prevents lysis of mouse/human hybrid cells by human complement. Scand J Immunol 1991; 34: 771.
8. Dalmasso AP, Vercellotti GM, Fischel RJ, et al. Mechanisms of complement activation in the hyperacute rejection of porcine organs transplanted into primate recipients. Am J Pathol 1992; 140: 1157.
9. White DJG, Langford GA, Cozzi E, Young VK. Production of pigs transgenic for human DAF: a strategy for xenotransplantation. Xenotransplantation 1995; 2: 213.
10. Schmoeckel M, Nollert G, Shahmohammadi M, et al. Prevention of hyperacute rejection by decay accelerating factor in xenogeneic perfused working hearts. Transplantation 1996; 69: 729.
11. Waterworth PD, Cozzi E, Tolan MJ, et al. Pig-to-primate cardiac xenotransplantation and cyclophosphamide therapy. Transplant Proc 1997; 29: 899.
12. White DJG. HDAF transgenic pig organs: are they concordant for human transplantation? Xeno 1996; 4: 50.
13. Cozzi E, White DJG. The generation of transgenic pigs as potential organ donors for humans. Nature Med 1995; 1: 964.
14. Cozzi E, Langford GA, Wright L, et al. Comparative analysis of human DAF expression in tissues of transgenic pigs and man. Transplant Proc 1995; 27: 319.
15. Rosengard AM, Cary NRB, Langford GA, et al. Tissue expression of human complement inhibitor, decay-accelerating factor, in transgenic pigs. Transplantation 1995; 59: 1325.
16. Lower RR, Shumway NE. Studies on the orthotopic homotransplantation of the canine heart. Surg Forum 1960; 11: 18.
17. Stark JH, Smit JA, Gridelli B. Sensitivity of baboon lymphocytes to cyclosporin A and FK 506: relative resistance of alloactivated cells to CyA. Transpl Int 1994; 7: 372.
18. Euler HH, Schroder OJ, Harten P, Zeuner RA, Gutschmidt HJ. Treatment-free remission in severe systemic lupus erythematosus following synchronization of plasmapheresis with subsequent pulse cyclophosphamide. Arthritis Rheum 1994; 12:1784.
19. Boumpas DT, Austin HA, Vaughn EM, et al. Controlled trial of pulse methylprednisolone versus two regimens of pulse cyclophosphamide in severe lupus nephritis. Lancet 1992; 340: 741.
20. Fukushima N, Gundry SR, Matsumiya G, Bouchart F, Zuppan C, Bailey LL. Histological findings in heart grafts after orthotopic pig to baboon cardiac transplantation. Transplant Proc 1996; 28: 788.
21. Cozzi E, Yannoutsos N, Langford GA, Pino-Chavez G, Wallwork J, White DJG. Effect of transgenic expression of human decay accelerating factor on the inhibition of hyperacute rejection of pig organs. In: Cooper DKC, Kemp E, Platt JL, White DJ, eds. Xenotransplantation: the transplantation of organs and tissues between species, 2nd ed. Berlin, Heidelberg, New York: Springer, 1997: 665.
22. Hasan RI, van den Bogaerde J, Forty J, Wright L, Wallwork J, White DJ. Prolonged survival of hamster to rat heart xenografts with cyclophosphamide therapy. Transplant Proc 1992; 24: 517.
23. Calne RY, White DJG, Rolles K, Smith DP, Herbertson BM. Prolonged survival of pig orthotopic heart grafts treated with cyclosporin A. Lancet 1978; 1: 1183.
24. Müller-Werdan U, Klein D, Zander M, Werdan K, Hammer C. Beating neonatal rat cardiomyocytes as a model to study the role of xenoreactive natural antibodies in xenotransplantation. Transplantation 1994; 58: 1403.
25. Autenrieth A, Müller-Werdan U, Koidl B, Werdan K, Hammer C. Different action of xenoreactive natural and immune antibodies on functioning cardiomyocytes. Transplant Proc 1996; 28: 576.
26. Hammer C. Evolution: its complexity and impact on xenotransplantation. In: Cooper DKC, Kemp E, Platt JL, White DJG, eds. Xenotransplantation: the transplantation of organs and tissues between species, 2nd Edition. Berlin, Heidelberg, New York: Springer, 1997: 716.
27. Sadeghi AM, Robbins RC, Smith CR, et al. Cardiac xenotransplantation in primates. J Thorac Cardiovasc Surg 1987; 93: 809.
28. 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.
29. Blakely ML, van der Werf WJ, Berndt MC, Dalmasso AP, Bach FH, Hancock WW. Activation of intragraft endothelial and mononuclear cells during discordant xenograft rejection. Transplantation 1994; 58: 1059.
30. Byrne GW, McCurry KR, Martin MJ, et al. Transgenic pigs expressing human CD59 and decay-accelerating factor produce an intrinsic barrier to complement mediated damage. Transplantation 1997; 63: 149.
31. Leventhal JR, John R, Fryer JP, Witson JC, et al. Removal of baboon and human antiporcine IgG and IgM natural antibodies by immunoadsorption. Transplantation 1995; 59: 294.
32. Cooper DKC, Human PA, Lexer G, et al. Effects of cyclosporine and antibody adsorption on pig cardiac xenograft survival in the baboon. J Heart Transplant 1988; 7: 238.
33. Roslin MS, Zisbrod Z, Burack JH, et al. 15-day survival in pig-to-baboon heterotopic cardiac xenotransplantation. Transplant Proc 1992; 24: 572.