0041-1337/00/7007-989/0TRANSPLANTATION Vol. 70, 989–998, No. 7, October 15, 2000Copyright © 2000 by Lippincott Williams & Wilkins, Inc. Printed in U.S.A.100

TransplantationT

RAPID COMMUNICATIONS

LIFE-SUPPORTING HUMAN COMPLEMENT REGULATOR DECAYACCELERATING FACTOR TRANSGENIC PIG LIVER XENOGRAFTMAINTAINS THE METABOLIC FUNCTION AND COAGULATION IN

THE NONHUMAN PRIMATE FOR UP TO 8 DAYS1

P. RAMIREZ,2,8 R. CHAVEZ,3 M. MAJADO,2 V. MUNITIZ,2 A. MUNOZ,4 Q. HERNANDEZ,2

C. G-PALENCIANO,2 G. PINO-CHAVEZ,5 M. LOBA,2 A. MINGUELA,2 J. YELAMOS,2 M. R. GAGO,2

A. S. VIZCAINO,2 H. ASENSI,2 M. G. CAYUELA,2 B. SEGURA,2 F. MARIN,2 A. RUBIO,2 T. FUENTE,2

R. ROBLES,2 F. S. BUENO,2 T. SANSANO,2 F. ACOSTA,2 J. M. RODRIGUEZ,2 F. NAVARRO,6

J. CABEZUELO,2 E. COZZI,5–7 D. J. G. WHITE,5–7 R. Y. CALNE,7 AND P. PARRILLA2

Liver Transplant Unit, Department of Surgery, University Hospital Virgen Arrixaca, Murcia, Spain;Transplant Unit. Department of Surgery, University Hospital of Wales, Cardiff. UK; School of Veterinary Medicine,

Murcia University, Spain; Imutran, Cambridge, UK; Department of Surgery, University of Montpellier,France; and Department of Surgery, University of Cambridge, Douglas House, Cambridge, UK

Background. It is not known whether the pig liver iscapable of functioning efficiently when transplantedinto a primate, neither is there experience in trans-planting a liver from a transgenic pigs expressing thehuman complement regulator human complementregulator decay accelerating factor (h-DAF) into a ba-boon. The objective of this study was to determinewhether the porcine liver would support the meta-bolic functions of non-human primates and to estab-lish the effect of hDAF expression in the prevention ofhyperacute rejection of porcine livers transplantedinto primates.

Methods. Five orthotopic liver xenotransplants frompig to baboon were carried out: three from unmodifiedpigs and two using livers from h-DAF transgenic pigs.

Findings. The three control animals transplantedwith livers from unmodified pigs survived for lessthan 12 hr. Baboons transplanted with livers fromh-DAF transgenic pigs survived for 4 and 8 days. Hy-peracute rejection was not detected in the baboonstransplanted with hDAF transgenic pig livers; how-ever, it was demonstrated in the three transplants

from unmodified pigs. Baboons transplanted with liv-ers from h-DAF transgenic pigs were extubated atpostoperative day 1 and were awake and able to eatand drink. In the recipients of hDAF transgenic piglivers the clotting parameters reached nearly normallevels at day 2 after transplantation and remainednormal up to the end of the experiments. In thesehDAF liver recipients, porcine fibrinogen was first de-tected in the baboon plasma 2 hr postreperfusion, andwas present up to the end of the experiments. Oneanimal was euthanized at day 8 after development ofsepsis and coagulopathy, the other animal arrested atday 4, after an episode of vomiting and aspiration. Thepostmortem examination of the hDAF transgenic liverxenografts did not demonstrate rejection.

Interpretation. The livers from h-DAF transgenicpigs did not undergo hyperacute rejection after ortho-topic xenotransplantation in baboons. When HAR isabrogated, the porcine liver maintains sufficient coag-ulation and protein levels in the baboon up to 8 daysafter OLT.

Research into xenotransplantation, stimulated by theshortage of organs for clinical use has been making signifi-cant progress in recent years (1). Although non-human pri-mates were initially regarded as potential donors, the pig iscurrently considered the most likely candidate as a source oforgans for medical application, due to advantages of size,ease of breeding, less risk of virus disease transmission, andethical considerations (2–4). The biology and immunopathol-ogy of heart and kidney xenotransplants have been studied inthe pig to non-human primate combination; extended sur-vival and improved control of hyperacute rejection (HAR)have been recently reported using hearts and kidneys fromtransgenic pigs expressing the human complement regulatordecay accelerating factor (hDAF) (5, 6). However, the expe-rience with xenotransplantation of pig livers into non-human

1 Supported by the “Consejeria de Sanidad y Consumo,” Comu-nidad Autonoma de Murcia, Spain, and CICYT (1FD97–0366); H.A.is supported by grant from the “Fundacion Seneca,” Consejeria deCultura, Comunidad Autonoma de Murcia, and J.Y. is supported bythe FIS.

2 Liver Transplant Unit, Department of Surgery, University Hos-pital Virgen Arrixaca.

3 Transplant Unit, Department of Surgery, University Hospital ofWales.

4 School of Veterinary Medicine, Murcia University.5 Imutran, Cambridge. UK6 Department of Surgery, University of Montpellier, France.7 Department of Surgery, University of Cambridge.8 Address correspondence to: P. Ramirez, MD, PhD, Unidad de

Trasplantes, Hospital Universitario Virgen Arrixaca, El Palmar,Murcia, Spain.

989

primates is limited (7) and the role of the various pathophys-iological components of xenograft rejection has not been de-termined in this model. With the exception of three experi-ments that lasted for a maximum of 3 days (8, 9), in mostreports the survival of liver xenografts from unmodified pigsto non-human primates did not exceed 12 hr. This is primar-ily due to the immediate onset of hyperacute rejection withcongestion, hemorrhage, and necrosis of the graft associatedwith severe coagulopathy. Makowka (10) performed the onlyclinical case of hepatic xenotransplantation of a pig liver in1994, in a patient with fulminant hepatitis who was submit-ted to preoperative plasmapheresis to remove the circulatingxenoantibodies. The graft was placed in a heterotopic posi-tion. The liver did not show any signs of metabolic activityand the patient died 30 hr after the transplant.

There is limited information on the ability of a pig liver tofunction efficiently when transplanted into a primate, andprevious in vitro studies suggested the incompatibility ofclotting factors (11). In addition to this, no information isavailable on the effect on hyperacute rejection of transplant-ing into primates a liver from a transgenic hDAF pig.

The objective of this study was to determine whether theporcine liver would support the metabolic functions of thenon-human primate and to establish the effect of hDAF ex-pression in modifying hepatic hyperacute rejection.

MATERIAL AND METHODS

All animal experiments were carried out according to the Code ofPractice For Scientific Procedures in Animals of the University ofMurcia, Spain according to recommendations of The Spanish Na-tional Committee of Xenotransplantation (12).

Animals. Five orthotopic liver xenotransplants from Large white-Landrace pig to baboon have been carried out. Three livers were fromunmodified specific pathogen free (SPF) pigs (5.5–6.5 kg) (cases 1 to3) and two were from an hDAF transgenic pig (8 and 10 kg) (cases 4and 5). The recipients were five baboons (Papio anubis) of 8–10 kg.

SPF pigs were obtained by hysterectomy of previously selectedand screened mothers. The piglets were kept under high biosecurityconditions (13). Large white/Landrace hDAF transgenic pigs (14, 15)(Courtesy of Imutran-Novartis, Cambridge UK) were used in thisstudy. Pigs with a high level of hDAF expression in the endothelium,arterial smooth muscle, and soft tissue of skin biopsies were selectedfor this study; skin biopsies were stained by immunoperoxidasemethod using a monoclonal anti-hDAF antibody (BRIC 216; IBGRLResearch Products, Elstree, UK). The blood groups in the five pairsof pig to baboon experiments were 0 to B in all cases.

Surgical procedure. The donor operation was performed with astandard technique using the University of Wisconsin Solution ascold preservation fluid, being the cold ischemic time less than 4 hr inthe five experiments. Anesthesia was induced with ketamine (1mg/kg) and propofol (1–2 mg/kg) followed by endotracheal intubationand muscular relaxation with vecuronium (0.1 mg/kg).

The five liver transplants reported here followed standard surgi-cal techniques for human orthotopic liver transplantation with se-quential end-to-end anastomoses of the supra-hepatic inferior venacava (IVC), infrahepatic IVC, portal vein, followed by reperfusion ofthe liver and declamping of the IVC, and finally end-to-side anasto-mosis of the hepatic artery to the infra-renal aorta using a conduit ofdonor aorta. The common bile duct was drained directly to theexterior with an indwelling catheter. The per-operative managementincluded central venous line, arterial line, and endotracheal tube,and anesthesia was induced with ketamine (1 mg/kg i.v.), propofol (2mg/kg i.v.), and vecuronium (0.1 mg/kg i.v.) and maintained withmidazolam (0.005 mg/kg per hour), ketamine (2 mg/kg/hr), vecuro-nium (0.1 mg/kg/hr), and fentanyl (1 mg/kg supplements) were used.

During the operation the animals received 3, 5, 4, 4, and 6 U,respectively, of compatible baboon total blood (1 U560 ml); tranex-amic acid was used intraoperatively at a dose of 1 mg/kg/hr duringthe hepatectomy and the anhepatic phase. Immunosuppression wasinduced by cyclophosphamide (CyP) on days 21 (40 mg/kg i.v.) andday 0 (20 mg/kg/i.v.), cyclosporine A (CyA) (day 21: 10 mg/kg i.v.; day0: 15 mg/kg i.v., before operation; and methylprednisolone (day 0: 1mg/kg i.v.).

Maintenance immunosuppression consisted of CyA, 10 mg/kg i.v.every 12 hr, changed after reestablishment of bowel transit to oralformulation (Neoral) 100 mg/kg every 12 hr. CyA levels were mea-sured every day by radioimmunoassay (RIA) to adjust the dose andreach target trough levels of 1000 ng/ml (16). Methylprednisolone (1mg/kg/i.v. every day) and CyP were given according to white bloodcells and platelets counts: In case 4: CyP was administered on day 2(30 mg/kg i.v.) and day 5 (15 mg/kg i.v.). Methylprednisolone 150 mg.was given i.v. on days 2, 3, and 4. In case 5, CyP was administeredon days 2 (10 mg/kg i.v.) and 4 (20 mg/kg i.v.).

Splenectomy was performed in all cases as a part of the immuno-suppression. Enteral Nutrition (Pentaset Pack Intensive, Nutricia,Zoetermeer, Holland) was given via nasogastric tube from day 1 in avolume of 150 ml (7.5 g protein/100 ml) plus 75 ml of water. Antibi-otic cover included Imipenem 500 mg i.v. 6 hourly, vancomicyn (25mg i.v./kg/12 hr) and amikacine (150 mg/24 hr i.v.).

During the postoperative period, the first transgenic transplant(OlxT4-hDAF1) received 80 ml of xenoantibody-free baboon plasmai.v. at days 4 and 5, and also 60 ml of washed, concentrated eryth-rocytes i.v. on days 5 and 6. In the second transgenic case (OlxT5-hDAF2) 120 ml of washed, concentrated erythrocytes were given onday 2, and 60 ml on day 3.

Xenoantibodies (Xab) were depleted from plasma according to amodification of the technique of Henry et al. (17) to diminish thepotential of xenoantibody mediated rejection induced by plasma.Plasma was obtained from 10% citrated baboon blood. A liver froman unmodified pig was explanted by standard surgical technique andperfused with nonheparinized cold 0.9 saline. The liver was perfusedex vivo by gravity with the plasma. Xab were determined before andafter perfusion to assure a clearance of at least 80%. Clotting factorsremained stable after this procedure (18). In cases 4 and 5, vitaminK was administered intravenously at a daily dose of 5 mg/kg.

Hematological and biochemical analysis. Blood samples weretaken during the dissection of the liver, the anhepatic phase andafter reperfusion of the graft at minutes 3, 15, 60, and 120. In theexperiment number three, blood samples were also taken at 4 and 8hr after reperfusion. In the experiment number 4 (hDAF), bloodsamples were taken daily for 8 days. In the experiment number 5(hDAF), blood samples were taken daily for 4 days. Biopsies of thedonor liver were taken before the implant and at 2 hr after reperfu-sion. Liver samples were also taken postmortem.

All the blood samples were investigated for white blood cells(WBC), hemoglobin, hematocrit, platelet count, coagulation studies(PT, APTT, fibrinogen, clotting factors), and IgM/IgG xenoantibod-ies. In the experiments number four and five (hDAF), daily bloodsamples were also analyzed for glucose, urea, creatinine. Na, K, P,Ca, uric acid, ALT, Alk. Phos., LDH, bilirubin, total proteins, albu-min, and protein electrophoresis. Prothrombin time (PT), activatedpartial thromboplastin time (APTT), and fibrinogen were analyzedin a photo-optical automatic coagulometer (Sysmex CA-1000, TOAMedical Electronics, Kobe, Japan). D-Dimer was measured by auto-mated latex enhanced immunoassay (IL Coagulation Systems ACLFutura Plus, Lexington, MA). Plasma samples were frozen for deter-mination of coagulation factors. Factors II, V, VII, VIII, IX, X, XI, andXII were determined by coagulometric techniques, by mixing thetested samples with human plasma specifically deficient in the factorto be measured. Platelets, leukocytes, erythrocytes, and hemoglobinwere measured in an automatic blood counter (Cell-Dyn 3500, Ab-bott, Abbott Park, IL). Agarose-gel high-resolution protein electro-phoresis (Beckman, Galway, Ireland) was used for the electro-

TRANSPLANTATION990 Vol. 70, No. 7

phoretic separation of serum proteins, this method can distinguishthe migration bands of porcine and primate fibrinogen, which are ofdifferent size (Fig. 2). Densitometry was performed by a standardtechnique (Helen Bioscience, Beaumont, TX). Xenoantibodies (XaB)were determined by indirect antiglobulin-hemaglutination of pigerythrocytes and by flow cytometry using pig lymphocytes as targets.Anti-pig lymphocytes IgM (xIgM) and IgG (xIgG) were measured byflow cytometry using anti-human IgM-FITC and IgG-FITC poly-clonal antibodies (Dako, Glostrup, Denmark). C3 and C4 plasmalevels were assayed by nephelometry (Beckman, Galway, Ireland),and CH50 by a standard haemolytic assay (19).

Immunohistochemistry. Liver biopsies were kept frozen at minus70°C until processed. Frozen sections were cut at 6 mm onto poly-1lysine- (Sigma Chemical Co., St. Louis, MO) coated slides and im-munostained using the avidin-biotin immunoperoxidase method forcomplement, immunoglobulins, fibrin, human-DAF, and counter-stained with H&E anti-human C3 mouse monoclonal antibodies(MoAb) (DAKO); rabbit anti-human C4 PoAb (DAKO); mouse Anti-human C5b9 (MAC) MoAb (Immunotech); mouse anti-human IgG(DAKO); mouse anti-human IgM (Immunotech); anti-human fibrin(Vector Laboratories, Burlingame, CA) and anti-pig pecam (Courtesyof Imutran-Novartis. Cambridge, UK). All the anti-human antibod-ies used in this experiment were previously tested to confirm cross-reaction with primate antigens. Formalin-fixed samples taken at theend of each experiment were processed and stained with hematoxilin& eosin.

RESULTS

Three control animals transplanted with livers from un-modified pigs survived for 2, 3, and 8 hr. The two baboonstransplanted with livers from hDAF transgenic pigs survivedfor 8 and 4 days, respectively.

In all five orthotopic liver xenotransplants (OlxT), therewas a dramatic fall of plasma anti-pig IgM and IgG titers 3min after reperfusion, accompanied by a fall in the levels ofC3, C4, and CH50 (hemolytic activity of complement) (Fig. 1).

Liver biopsies taken 2 hr after reperfusion showed severeedema and sinusoidal congestion in the three control cases,with destruction of vascular and sinusoidal endothelium andparenchymal hemorrhage. Immunohistology demonstratedendothelial deposits of IgG, IgM, C3, C4, and C5b-9 (20) (Fig.2). However, in the hDAF expressing transgenic livers, thehepatic architecture was preserved and no deposits of C3 orC5b-9 were found after reperfusion (Fig. 2). Deposits of im-munoglobulin and C4 were seen in the biopsies of the trans-genic livers. This suggests protection of the liver from thealternative complement pathway and incomplete activationof the classical pathway, as would be anticipated from anhDAF transgenic liver. In these cases, the prereperfusioncontrol biopsies showed expression of hDAF in the vascularand sinusoidal endothelium (Fig. 2).

The three control animals required continuous mechanicalventilation, did not produce any bile, developed increasingmetabolic acidosis, hyperkalemia, diffuse hemorrhage, andshock that conduced to cardiac arrest.

In the hDAF transgenic OlxT that survived for 8 days (case4), the hemodynamic conditions were stable at 12 hr aftersurgery, with no evidence of bleeding. Prothrombin time (PT)was of 27.2 sec and activated plasma thromboplastin time(APTT) was 65 sec (control 29 sec). There was good bileoutput through the drainage tube and good urine output.

The animal was successfully extubated and was awake andable to eat and drink despite sedation with continuous i.v.

infusion of Propofol (1/mg/kg/hr) and s.c. morphine (0.05 mg/kg/4 hr). Thirty-six hours after surgery, the bile outputdropped gradually to zero and there was an elevation of theaminotransferase (ALT), bilirubin, and lactic dehydrogenase(LDH) (Table 1a), with a PT of 15.8 sec and an APTT of 34 sec(control 29 sec).

A cholangiogram at this time demonstrated integrity of thebiliary system. Eco-Doppler imaging demonstrated the pa-tency of the hepatic artery, portal vein, and inferior venacava (IVC). At this stage, the onset of delayed acute vascularrejection was suspected and methyl prednisolone (150 mgi.v.) was given for 3 days. There was recovery of the bileproduction and the liver function tests improved progres-sively after this treatment started (Table 1a).

Ascites was detected from day 2 posttransplant and ortho-static edema was present from day 4, coinciding with a rel-ative hypoproteinemia, with levels of albumin of 1.8 and 2.2g/100 ml (Table 1a). At this stage (day 4), the renal functionwas preserved, with no proteinuria. Eco-Doppler demon-strated patency of the IVC and confirmed the presence ofascites (data not shown). The baboon received xenoantibody-free compatible baboon plasma, (120 ml in 48 hr). The edemaresolved and the volume of ascites reduced, although it didnot disappear completely.

The animal developed progressive anemia, with hemoglo-bin of 5.7 g/dl and hematocrit of 17% at postoperative day 5.An elevation of both conjugated and unconjugated bilirubin

FIGURE 1. Pre and posttransplant baboon serum levels of A,anti-pig lymphocytes IgG and IgM measured by flow cytom-etry using anti-human IgM-FITC and IgG-FITC polyclonalantibodies, expressed as mean fluorescent intensity (MFI). B,Plasma levels of C3 and C4 assayed by nephelometry. C, Com-plement haemolytic activity (CH50) expressed in relativeunits (R.U). The dotted lines represent the mean (SD) of agroup of 10 healthy baboons.

RAMIREZ ET AL.October 15, 2000 991

FIGURE 2. Histopathology of the pig liver after transplantation into baboon. (1) Hyperacute rejection (HAR) in a postmortemsample of a control animal. Wild-type pig liver shows HAR, characterized by massive hemorrhage, sinusoidal disruption, andhepatocyte vacuolisation. (H&E x10). (2) Immunoperoxidase staining of the same liver with anti-human C5b-9 shows depo-sition of membrane attack complex along blood vessels and sinusoids (x10). (3) Anti-human C3 immunoperoxidase staining ina postmortem sample of an hDAF transgenic pig liver after transplantation. The absence of C3 deposits demonstrated theabrogation of the alternative pathway of the complement cascade (x10). (4) Immunoperoxidase staining with anti-human C4in a postmortem sample of an hDAF transgenic pig liver after transplantation. Deposits of C4 demonstrate the initialactivation of the classical pathway of the complement cascade; this incomplete pathway is not affected by the presence ofhDAF (x10). (5) Anti-human C5b-9 immunoperoxidase in a post mortem sample of an hDAF transgenic pig liver aftertransplantation. The absence of deposits of C9 demonstrates the abrogation of the alternative pathway and of the postC4pathway of the complement cascade (x10). (6, 7) Immunoperoxidase staining of prereperfusion control biopsies of an hDAFtransgenic liver, using BRIC-216, shows expression of hDAF in the vascular and sinusoidal endothelium (x5 and x20 respec-tively). (8) Absence of rejection in an hDAF transgenic pig liver, 8 days after transplantation into a baboon (case 4). Lobararchitecture is preserved, with sinusoidal integrity and absence of lymphocyte infiltrate. A small focus of ischemic necrosisis visible in the left side of the image (H&E x10). (9) Absence of rejection, with normal liver morphology in an hDAF transgenicpig liver 4 days after transplantation (case 5). The lobular architecture of the graft is preserved, with integrity of the sinusoidsand liver plates, absence of haemorrhage or white blood cell infiltrate. (H&E x5).

TRANSPLANTATION992 Vol. 70, No. 7

was noted, together with a negative direct antiglobulin test(Coombs’). ABO-matched packed washed red baboon cells(120 ml) were transfused at this stage.

The clinical condition of the recipient remained stable up today 7, when there was pyrexia (39.5°C), anorexia, derange-ment of liver function tests (LFTs), and a fall in fibrinogenlevels (Table 1). The bile output was unchanged. At the endof postoperative day 8, temperature spiked again to 40°C,and the baboon developed cutaneous bruises and bleeding

through puncture sites, dyspnea, and oliguria. The plateletcount was 34,000, with a fall of fibrinogen levels to 68 mg/100ml and an elevation of DD to 4992 ng/ml. The baboon waseuthanaized at this stage.

In the hDAF transgenic OlxT which survived for 4 days(case 5), the hemodynamic conditions were stable at 12 hrafter surgery, with no evidence of bleeding, prothrombin time(PT) of 19.1 sec and a normal activated plasma thromboplas-tin time (APTT) of 26 sec.

TABLE 1. Hematology, biochemistry, and coagulation before and after orthotopic liver xenotransplantation of two hDAFtransgenic pig livers into baboons

Piglet Baboon 1 Day 2 Day 3 Day 4 Day 5 Day 6 Day 7 Day 8 Day

a, Case 4Glucose (mg/dl) 104 142 221 121 44 94 128 106 124 70Creatinine (mg/dl) 1.6 1.2 1 1.8 1.3 2 2.2 1.7 1.5 1.7Uric acid (mg/dl) 0 0 0 0.5 0.7 0.7 0.8 0.5 1.5 2Urea (mg/dl) 13 31 24 27 35 79 125 129 130 148Calcium (mg/dl) 11.3 10.5 8.6 7.3 7.6 7.7 8.2 8.7 8.8 8.9Phosphate (mg/dl) 10.2 3.9 6.8 6 5.5 5.4 5.8 3.5 4.5 6.6Proteins (gr/dl) 4.8 7.5 4.1 3.5 4.1 3.8 3.4 3.9 3.9 3Albumin (gr/dl) 3.7 4.1 1.9 1.8 2.1 2.2 1.9 2.2 2.3 1.7Total bilirubin (mg/dl) 0.1 0.1 0.3 1.2 7.6 12.5 11.7 14.2 18.6 18.9Conjugated bilirubin mg/dl 0 0 0.01 0.03 3.8 8.5 8.35 10.5 12 11.9LDH (U/liter) 909 198 1,798 14,320 12,250 13,430 7,300 5,530 5,500 9,810AST (U/liter) 40 21 445 3,210 2,910 2,790 1,070 210 920 4,920GGT (U/liter) 49 67 44 49 71 64 59 68 79 87Alk. Phos. (U/liter) 1,063 3,838 1,692 2,623 2,757 2,355 1,559 1,409 1,200 824ALT (U/liter) 24 6 38 180 170 160 106 103 81 93Bile output (ml/24 hr) 30 31.5 0 0.5 7.7 8.3 5.8 14.5Triglicerides (mg/dl) 49 44 20 18 34 60 66 56 116 110Cholesterol (mg/dl) 75 94 54 39 33 28 40 57 66 46Hematocrit (%) 41.9 36.8 30 29.8 28.6 27.8 16.9 18.3 24.4 23.6Hemoglobin (g/dl) 12 11.5 9.7 9.7 9.5 9.6 5.7 6.1 7.3 8WBC 20,000 8,600 7,800 4,500 3,800 3,800 5,500 6,500 5,500 3,700Platelets 356,000 300,000 112,000 57,000 77,000 63,000 38,000 46,000 59,000 34,000Prothrombin (sec) 12 12 27.2 19 15.8 19.3 17.4 22.6 22.5 29.1APTT (sec) 12.8 24 39.7 34 21 20.1 18.8 25.4 30.9 35.3Fibrinogen (mg/dl) 234 273 99 181 431 460 437 408 173 68.8

b, Case 5Glucose (mg/dl) 114 111 137 37 287Creatinine (mg/dl) 1.2 1.3 2.8 5.1 4.9Uric acid (mg/dl) 0 0 0.7 0 0Urea (mg/dl) 12 22 42 102 118Calcium (mg/dl) 10.9 9.2 6.5 5.4 6.1Phosphate (mg/dl) 10.3 1.1 5.4 9.6 6.3Proteins (g/dl) 4.5 6.9 3.4 4.1 2.9Albumin (g/dl) 3.7 3.3 1.3 1.5 1.2Bilirubin (total) (mg/dl) 0.1 0.1 0.9 4.3 5.4Conjugated bilirubin (mg/dl) 0 0.01 3.3 4.3LDH (U/liter) 927 983 22,040 25,070 24,860AST (U/liter) 47 28 2,210 4,050 835GGT (U/liter) 57 44 21 29 37Alk. Phos. (U/liter) 1,106 1,188 1,450 1,465 1,032ALT (U/liter) 28 18 147 187 128Bile output (ml/24 hr) 94 23 30Triglicerides (mg/dl) 49 57 125 165 95Cholesterol (mg/dl) 75 76 48 41 28Hematocrit (%) 38.8 31.2 23.2 16.2 21Hemoglobin (gr/dl) 11.7 9.3 7.4 5.6 6.6Leukocytes 16,000 7,000 7,000 1,480 1,300Platelets 296,000 443,000 31,000 44,000 50,000Prothrombin (sec) 11.9 14.7 19.1 13.1 11.9APTT (sec) 14.7 31.4 26.8 22.4 16.1Fibrinogen (mg/dl) 254 267 22 512 693

RAMIREZ ET AL.October 15, 2000 993

This animal was also extubated and was awake and able toeat and drink under sedation with continuous i.v. infusion ofPropofol (1/mg/kg/hr) and s.c. morphine (0.05 mg/kg/4 hr).There was good bile output trough the drainage tube andgood urine output. The bile output and the coagulation pa-rameters remained within the normal range throughout theexperiment (Table 1b).

The albumin level was 1.5 g/100 ml at day 2, withoutascites in this case. This recipient also developed anemia,with hemoglobin of 5.6 g/dl and hematocrit of 16% at postop-erative day 2. ABO matched packed washed red baboon cells(180 ml) were transfused. The clinical conditions remainedstable up to the beginning of day 4, when the animal arresteddue to aspiration after an episode of vomiting during naso-gastric drug administration.

Postmortem histological examination of the hDAF trans-genic liver xenografts demonstrated that neither liver hadbeen rejected. In case 4, there was evidence of some patchyfocal ischemia and in case 5 the morphology of the liver wasentirely normal (Fig. 2). In case 4 a bilateral bronchopneu-monia was demonstrated and in case 5 pulmonary bron-choaspiration was confirmed.

Coagulation and protein profile. All the transplanted an-imals showed consumption of clotting factors after reperfu-sion of the graft. However, in the hDAF transgenic pig liver

grafts, clotting factors recovered progressively. This was as-sociated to the appearance of porcine type fibrinogen (Fig. 3)that reached a level of above 431 mg/100 ml, at day 4 in thefirst transgenic case (Table 1a) and a level of 512 mg/100 mlat day 2 in the second transgenic case (Table 1b).

The clotting factors recovered progressively in both therecipients, except for factor II that did not recover above 30%in both cases, with no petechial haemorrhages or any otherevidence of hemorrhagic diathesis in either case. However,case 4 developed disseminated intravascular coagulation(DIC) associated with sepsis on day 8 that led to the termi-nation of the experiment (Table 2).

Baboon fibrinogen disappeared from plasma at day 3 andat 2 hr after liver reperfusion in cases 4 and 5, respectively(Fig. 3) whereas baboon a2-macroglobulins and transferrinwere present up to the end of the experiment in both cases.Porcine fibrinogen was first detected in the baboon plasma at2 hr postreperfusion and was present up to the end of theexperiments.

The levels of baboon C3 and C4 were negligible at 6 hrafter reperfusion in both cases 4 and 5 and never recovered,however, the hemolytic complement activity (CH50) recov-ered to 64.5% at day 6 in case 4 and 82.5% at day 2 in case 5.This suggests the presence of newly synthesized porcine com-plement components (Fig. 1).

FIGURE 3. Pre- and posttransplant protein profile in two donor pigs and two baboons transplanted with hDAF transgenic piglivers. A, Agarose-gel high-resolution protein electrophoresis (Beckman, Galway, Ireland) was used for the electrophoreticseparation of serum proteins. In both cases, a switch from baboon-type to porcine-type fibrinogen can be appreciated. B,Posttransplant plasma levels of porcine versus baboon-type fibrinogen expressed as a percentage of the total proteinconcentration in the plasma. There is early appearance of porcine type fibrinogen in both recipient baboons. Densitometrywas performed by a standard technique (Helen Bioscience, Beaumont, TX).

TRANSPLANTATION994 Vol. 70, No. 7

DISCUSSION

There is very limited experience of porcine liver xenotrans-plantation into primates (8, 9, 20–22), there is only oneclinical case reported in the literature (10) and there is noprevious experience using genetically modified pigs as liverdonors. Only three non-human primate cases have been re-ported to survive more than 3 days; one of them, by Calne in1968 (8) survived for 72 hr using corticoids as immunosup-pressive therapy, and died due to a respiratory infection. Theother two cases reported by Powelson et al. (9) 25 years later,survived 72 and 75 hr. In those two cases, the recipients hadprevious adsorption of xenoantibodies by extracorporeal per-fusion of their blood through the liver from another pig. Atthe end of those experiments, one of the livers showed cen-trilobular necrosis and the other was reported as normal. It isremarkable that in those three liver xenotransplants, the pigliver was capable of maintaining the blood coagulation of theprimate during the first few days after transplantation. Inour control experiments, using livers from unmodified pigs,histology indicated hyperacute rejection, with congestion andhemorrhage at 2 hr after reperfusion, followed by rupture ofthe endothelium, hemorrhage, intravascular thrombosis, andnecrosis in the post mortem samples (Fig. 2). Immunohisto-chemistry demonstrated deposits of IgG, IgM, C3, C4, andC5b-9, which confirmed the diagnosis of hyperacute rejection(20, 23).

Cold ischemic times were kept for fewer than 4 hr in all theexperiments, to minimize the impact of ischemia and reper-fusion; this time is well within the safe limits of up to 24 hrfor human liver allotransplantation when UW preservationsolution is utilized (24). The role of ischemia-reperfusion (I/R)injury in the development of HAR is a question relevant to allxenotransplantation experiments. Common pathways of en-dothelial and complement activation in ischemia/reperfusionand HAR are likely to interact in all xenotransplants and itmight be difficult to distinguish both entities when taken outof clinical context. Deposits of C3 and C5b-9 are a feature ofboth I/R and HAR. C4 deposits are not notorious in I/R notassociated with transplantation (25, 26), but are a constantfeature of all xenografts, independently of the presence ofHAR; therefore, the immunohistological finding of comple-ment deposits in the a xenograft is of difficult interpretationon its own. However, the presence of early endothelial dis-ruption with fibrin deposition, thrombosis, and hemorrhageis a feature of HAR that is not commonly seen in reperfusioninjury.

In contrast, in the liver transplants performed with hDAFexpressing porcine organs, with similar ischemic times, theliver architecture was preserved after reperfusion showingdeposits of C4, but not of the terminal components C5b-9(Fig. 2). This confirms the capacity of hDAF expressed by theliver graft to block the activation of the complement cascadeat the point of formation of C3 convertases, and thereforeprevent hyperacute rejection (27, 28), as previously reportedin kidney and heart xenotransplantation (5, 6).

Control cases 1 and 2 were found to have higher preoper-ative levels of xenoantibodies (Fig. 1), which may influencethe intensity of HAR. However, in these discordant xeno-transplants, the preoperative antibody levels of the recipi-ents were sufficient to produce similar activation of the clas-sical pathway of complement in all the cases (interrupted at

TA

BL

E2.

Pig

tob

aboo

nli

ver

xen

otra

nsp

lan

tati

on

PT

AP

TT

FII

VV

IIX

VII

IIX

XI

XII

aP

re-o

p12

/14

24/3

127

3/26

773

/87

90/5

216

3/17

370

/45

237/

9145

/40

58/2

930

6/21

0R

eper

f.25

.8/2

9.1

31/n

o13

7/54

41/1

650

/19

100/

6536

/10

77/2

017

/10

34/5

169/

63D

ay3

16/1

221

/16

431/

693

28/1

78/

692

/80

21/1

333

/20

13/9

33/5

141/

61D

ay4

19/1

0834

/16

460/

693

10/2

521

7/18

818

/93

24/6

338

9/20

913

2/21

076

/51

335/

210

Day

517

1943

713

271

3229

409

154

113

437

Day

622

.625

408

823

619

1624

311

110

933

5D

ay7

2935

6810

3128

1161

5939

90

bP

re-o

p13

.2/1

1.5/

15.3

29/3

3/35

250/

198/

198

83/7

5/59

88/5

6/49

198/

91/6

679

/87/

8122

6/91

/66

31/2

7/30

50/2

6/42

ND

/ND

/185

Rep

erf.

26.1

/26.

9/.

60.

150/

150/

143

47/.

20/4

919

/11/

1714

/3/1

186

/5/1

920

/14/

291/

0/8

2/0/

115/

0/20

ND

/ND

/113

a,C

oagu

lati

onfa

ctor

saf

ter

orth

otop

icli

ver

xen

otra

nsp

lan

tati

onof

two

hD

AF

tran

sgen

icpi

gli

vers

into

babo

ons.

Th

ecl

otti

ng

fact

ors

reco

vere

dpr

ogre

ssiv

ely

inbo

thca

ses

exce

ptfa

ctor

IIth

atdi

dn

otre

cove

rm

ore

than

30%

.T

her

ew

asn

oev

iden

ceof

hem

orrh

agic

diat

hes

isin

both

tran

spla

nte

dba

boon

s.[A

llth

eva

lues

are

expr

esse

das

perc

enta

ges

exce

ptfi

brin

ogen

(mg/

100

ml)

,PT

(sec

),an

dA

PT

T(s

ec).

Th

eva

lues

ofth

etw

oex

peri

men

tsar

ese

para

ted

bysl

ash

es(c

ase

4/ca

se5)

].b,

Coa

gula

tion

fact

ors

afte

ror

thot

opic

live

rxe

not

ran

spla

nta

tion

ofth

ree

un

mod

ifie

dpi

gli

vers

into

babo

ons.

[All

the

valu

esar

eex

pres

sed

aspe

rcen

tage

sex

cept

fibr

inog

en(m

g/10

0m

l),P

T(s

ec),

and

AP

TT

(sec

).N

D,N

otdo

ne.

Th

eva

lues

ofth

eth

ree

expe

rim

ents

are

sepa

rate

dby

slas

hes

(cas

e1/

case

2/ca

se3)

].

RAMIREZ ET AL.October 15, 2000 995

C4 level in the hDAF grafts). Control case 3 developed HARwhereas the transgenic hDAF cases did not, with similarantibody levels (Figs. 1 and 2).

The survival of the recipients, awake and conscious andtolerating oral nutrition, for 4 and 8 days has allowed theanalysis of the synthetic function of a porcine liver in aprimate recipient. We demonstrated that clotting factors syn-thesized by the porcine liver have functioned in vivo in pri-mate recipients for up to 8 days.

In the liver xenografts performed in this study, there wasearly consumption of clotting factors after reperfusion of thegraft, however, in the hDAF grafts, the clotting factors weresubsequently replaced by the newly synthesized proteins pro-duced by the porcine liver. Most clotting factors reachednormal levels. Circulating porcine fibrinogen was demon-strated in the baboon from the first postoperative day, withcapacity to participate in the control of hemostasis, as sug-gested by the absence of bleeding.

In addition, returning of complement activity (CH50) inthe absence of primate complement components C3 and C4 isstrongly suggestive of the presence of newly synthesized por-cine complement in the recipient serum.

Previous experiences with liver xenotransplants fromhamster to rat have established that the synthesis of proteinsby the donor liver is determined by its genetic configuration,therefore the recipient progressively acquires the proteinprofile of the donor species (29, 30). The rate of this phenom-enon depends on the half-life of the native proteins. In theseexperiments, baboon fibrinogen disappeared early after liverreperfusion, whereas baboon a2-macroglobulins and trans-ferrin remained to the end of the experiments. Porcine fibrin-ogen detected in the baboon’s plasma after transplantationindicated the synthetic function of the newly transplantedxenograft and is consistent with the acquisition by the ba-boon of a protein profile that resembles the pig (Fig. 3).

Fibrinogen levels were higher than the normal adult por-cine levels of 100 to 400 mg/100 ml (31). Whether this phe-nomenon is due to an acute phase reaction or there is afailure in the autoregulation mechanisms of fibrinogen is anaspect that deserves further investigation.

Even though comparative studies demonstrate a substan-tial physiological compatibility between the porcine and pri-mate species (32), it is surprising that in these cross-speciestransplants the liver xenograft was able to provide life-sup-porting synthetic function for up to 8 days, given the largenumber of metabolically important proteins produced in theliver.

Progressive anemia and hypoalbuminemia were the majorclinical problems detected after day 2 in the surviving ani-mals. The anemia might be multifactorial, with poor synthe-sis, increased losses, an adverse effect of immunosuppressivedrugs and finally, a theoretical hemolytic component withinvolvement of pig complement. This component could not beclearly demonstrated in this study because the complexity ofthe experiment obscured the indicators of hemolytic anemia.The participation of immune complexes was excluded as acause of hemolytic anemia (negative Coombs’ test).

Low albumin levels (Table 1, a) may have conditioned theclinical picture of transient ascites and edema. In the absenceof albuminuria, this may be due to inadequate synthesis ofalbumin by the porcine liver and/or to hemodilution. It isunlikely that hypoalbuminemia was solely related to liver

dysfunction in the context of sufficient synthesis of otherproteins such as fibrinogen and clotting factors. Interest-ingly, albumin and cholesterol levels found in the recipientwere within the normal range for the pig (33, 34). Moreexperiments, with longer survival will be necessary to inves-tigate whether this hypoalbuminemia represents a case ofmetabolic incompatibility as suggested by Starzl et al. (35) inOlxT from baboon to human.

In the first baboon transplanted with an hDAF pig liver, onpostoperative day 2, the bile output stopped, associated toderangement of LFTs. After confirmation of integrity of allvascular and biliary structures by eco-Doppler and cholan-giogram, this was interpreted as the clinical expression ofdelayed vascular rejection, and was successfully treated withboluses of methylprednisolone and an increment in the doseof CyP, with recovery of the bile production and improvementof liver function tests (Tables 1 and 2). In the days thatfollowed this episode, there was a picture of progressive cho-lestasis (Tables 1, a and b), with elevation of bilirubin andLDH in the presence of a reasonable hepatic function indi-cated by the animal been awake, synthesizing clotting fac-tors, and detoxifying the Propofol administered for sedation.

This cholestasis might represent a degree of xenograftrejection, although the histology of the liver did not showfeatures of DXR (Fig. 2). In this case, a picture of DIC,attributed to septicemia appeared at postoperative day 7.DIC has been previously described in renal and cardiac xe-notransplantation from pig to primate, associated to delayedxenograft rejection (DXR) (36) and may be associated intheory to primary graft nonfunction or vascular catastrophe,delayed rejection, or sepsis. In this experiment, primary non-function was not the case, after several days of liver functionin a conscious, alert animal. Anastomotic complications wereexcluded by eco-Doppler and by the postmortem examina-tion. The histology of the liver showed a reasonably pre-served hepatic architecture (Fig. 2). Some areas of focal ne-crosis may be related to uneven reperfusion of the graft dueto either operative trauma of a very delicate graft or to somedegree of reperfusion injury, or theoretically, to vascularrejection, although there are no histological features of rejec-tion in this graft. Pyrexia at day 7 and the postmortemfinding of pneumonia, even with negative blood cultures,suggest that sepsis may have had an influence in the onset ofDIC.

The results of this study demonstrate that the expressionof hDAF in these grafts has been able to prevent HAR pro-duced by the activation of primate complement in the graft.The porcine liver, expressing both hDAF and porcine comple-ment regulators is protected against complement, initially ofprimate origin and later, in theory, against porcine comple-ment synthesized by the graft. This dual complement protec-tion of the liver associated with the switch on the comple-ment phenotype may further protect the xenogeneic liveragainst the attack of complement. This situation is quanti-tatively unique to the liver, which is the main source ofcomplement and differs from other xenografts such as heartor kidney (29). Experiments are underway to investigatewhether hDAF expression will consistently protect the liverxenograft against HAR, as in the case of cardiac and renalxenotransplantation (5, 6).

Clinical trials of xenotransplantation of porcine organs andtissues are considered nowadays (37) and the results of this

TRANSPLANTATION996 Vol. 70, No. 7

study raise the question of whether the use of porcine liverswill become a viable clinical alternative for the managementof acute liver failure.

In a recent report of a clinical trial by Levy et al. (38)transgenic porcine livers (hCD55/hCD59) were extracorpore-ally perfused in two patients with acute liver failure for 6.5and 10 hr before transplantation. The procedure proved to besafe, and there was a discrete improvement of liver functiontests during the few hours of perfusion. There was a fall ofammonium levels in one case, bilirubin improved in bothcases, and prothrombin time improved discretely in the casewith ammonium elevation.

In our study, the survival of two baboons with orthotopicporcine liver grafts for 4 and 8 days, extubated, and awake,with demonstrable liver function, suggests that this alterna-tive may be more effective than extracorporeal perfusion.

A number of conditions such as fulminant hepatitis, poi-soning with hepatotoxic agents and primary nonfunction of aliver graft require an urgent transplant to restore the hepaticfunction; delay may result in rapid development of multior-gan failure with accelerated metabolic acidosis, hyperkale-mia, hypoglucemia, renal failure, coagulopathy, sepsis, andirreversible CNS damage due to encephalopathy and cerebraledema (39). Removal of the necrotic organ may be beneficialin some of those cases (40). In this context, an hDAF pig liverxenograft might be considered as a short-term bridge graftwhile a human allograft becomes available.

Acknowledgments. The authors thank C. Lagares, J. Pastor, V.Aroca, M. D. Martinez, M. D. Palazon, and M. M. Rodriguez fortechnical assistance.

REFERENCES

1. Cooper DKC. Introduction. In: Cooper DKC, Kemp E, Platt JL,White DJG, eds. Xenotransplantation, 2nd ed. Springer, Ber-lin: 1997; 2.

2. Cooper DKC, Ye Y, Rolf L Jr, et al. The pig as potential organdonor for man. In: Cooper DKC, et al., ed. Xenotransplanta-tion. Heidelberg: Springer, 1991, 481.

3. Ye Y, Niekrasz M, Kosanke S, et al. The pig as a potential organdonor for a man: a study of potentially transferable diseasefrom donor pig to recipient man. Transplantation 1994; 57:694.

4. Paridis K, Langford G, Long Z, et al. Search for cross-speciestransmission of porcine endogenous retrovirus in patientstreated with living pig tissue. Science 1999; 285: 1236

5. Zaidi A, Schmoeckel M, Bhatti F, et al. Life-supporting pig-to-primate renal xenotransplantation using genetically modifieddonors. Transplantation 1998; 65 (12): 1584.

6. Schmoeckel M, Bhatti FN, Zaidi A, et al. Orthotopic heart trans-plantation in a transgenic pig-to-primate model. Transplanta-tion 1998; 65 (12): 1570.

7. Lambrigts D, Sachs DH, Cooper DK. Discordant organ xeno-transplantation in primates: world experience and current sta-tus. Transplantation 1998; 66 (5): 547.

8. Calne RY, White HJO, Herbertson BM, et al. Pig to baboon liverxenografts. Lancet 1968; 1: 1176.

9. Powelson J, Cosimi AB, Austen W, et al. Porcine to primateorthotopic liver transplantation. Transplant Proc 1994; 26:1353.

10. Makowka L, Wu GD, Hoffman A, et al. Immunohistopathologiclesions associated with the rejection of a pig to human liverxenograft. Transplant Proc 1998; 26: 1074.

11. Lin SS, Weidner BC, Byrne GW, et al. The role of antibodies inacute vascular rejection of pig-to-baboon cardiac transplants.

J Clin Invest 1998; 101: 1745.12. Matesanz R, Miranda B. Recommendations of the National Com-

mittee of Xenotransplantation. Ministerio de Sanidad (Spain),Madrid, 1999.

13. Munoz A, Pallares FJ, Ramis G, et al. Surgical procedure forspecific pathogen free piglet production by modified terminalhysterectomy. Transplant Proc 1999; 31: 2627.

14. White DJ, Yannoutsos-N. Production of pigs transgenic for hu-man DAF to overcome complement-mediated hyperacute xeno-graft rejection in man. Res Immunol 1996; 147 (2): 88.

15. Cozzi E, Tucker AW, Langford GA, et al. Characterisation of pigstransgenic for human decay accelerating factor. Transplanta-tion 1997; 64: 1383.

16. Gridelli B, et al. Experimental concordant liver xenotransplan-tation in non-human primates. In: Xenotransplantation, 2nded. Cooper DKC, Kemp E, Platt JL, White DJG, eds. Springer,Berlin: 1997; 323.

17. Henry M, Han L, Davies E, Sedmak D, Ferguson R. Antibodydepletion prolongs xenograft survival. Surgery 1994; 115: 355.

18. Majado M, Ramirez P, Minguela A, et al. Normal CoagulationParameters after pig livers and kidneys perfusion with humanplasma, intended to xenoantibodies depletion. Transplant Proc1999; 31 (7): 2834.

19. Hudson L, and Hay FC. Inmunologia practica. In: Hudson L,Hay FC, eds. Barcelona: JIMS, 1979, 147.

20. Ramirez P, Chavez R, Majado M, et al. Study of xenograft rejec-tion in a model of liver xenotransplantation from unmodifiedpig to primate. Transplant Proc 1999; 31 (7): 2814.

21. Calne RY, Davis DR, Pena JR, et al. Hepatic allografts andxenografts in primates. Lancet 1970; 1: 103.

22. Luo Y, Kosanke S, Mieles L, et al. Comparative histopathology ofhepatic allografts and xenografts in the nonhuman primates.Xenotransplantation 1998; 5: 197.

23. Dalmasso AP, Vercellotti GM, Fischel RJ, Morton B, Bach FH,Platt JL. Mechanism of complement activation in the hyper-acute rejection of porcine organs transplanted into primaterecipients. Am J Pathol 1992; 140, 5: 1157

24. Chavez-Cartaya R, Rasmussen A, Tokat Y, Jamieson NV. Effectof preservation time on early graft function and outcome oforthotopic liver transplants. Transplant Proc 1995; 27 (1): 724.

25. Chavez-Cartaya R, Pino DeSola G, Wright L, Jamieson NV,White DJG. Regulation of the complement cascade by solublecomplement receptor type I. Protective effect in experimentalliver ischaemia and reperfusion. Transplantation 1995; 59:1047.

26. Vakeva A, Morgan BP, Tikannen I, Helin K, Laurila P, Meri S.Time course of complement activation and inhibitor expressionafter ischaemic injury of rat myocardium. Am J Pathol 1998;144–6: 1357.

27. Cozzi E, White DJG. The generation of transgenic pigs as poten-tial organ donors for humans. Nat Med 1995; 1: 964.

28. Cozzi, E, Yannoutsos G, Langford G, Pino-Chavez G, WallworkJ, White DJG. Effect of transgenic expression of human decayaccelerating factor on the inhibition of hyperacute rejection ofpig organs. In: Xenotransplantation. Cooper DKC, Kemp E,Platt JL, White DJG, eds. Berlin: Springer, 1997; 665.

29. Celli S, Valdivia LA, Fung JJ, Kelly RH. Early recipient-donorswitch of the complement type after liver xenotransplantation.Immunol Invest 1997; 26 (5–7): 589.

30. Valdivia LA, Lewis JH, Celli S, et al. Hamster coagulation andserum protein in rat recipients of hamster xenografts. Trans-plantation 1993; 56, 2: 489.

31. Dodds WJ. Hemostasis (chapter 10). (Appendix 8, Blood analysisreference values in large animals). In: Kaneko JJ, ed. Clinicalbiochemistry of domestic animals. San Diego: Academic Press,1997; 891.

32. Kobayashi T, Taniguchi S, Ye Y, Niekrasz M, Nour B, CooperDKC. Comparison of bile chemistry between humans, baboons,

RAMIREZ ET AL.October 15, 2000 997

and pigs: Implictions for clinical and experimental liver xeno-transplantation. Lab Anim Science 1998; 48 (2): 197.

33. Munitiz V, Ramırez P, Hernandez Q, et al. Hematological andhepatic function profile comparison between pig and baboon inan orthotopic liver xenotransplantation model. TransplantProc 1999; 31: 2641.

34. Lampreave F, Gonzalez N, Martinez S, et al. Characterization ofthe acute phase serum protein response in pigs. Electrophore-sis 1994; 15: 672.

35. Starzl TE, Fung J, Tzakis A, et al. Baboon-to-human liver trans-plantation. Lancet 1993; 1: 65.

36. Ierino FL, Kozlowski T, Siegel JB, et al. Disseminated intravas-cular coagulation in association with the delayed rejection ofpig-to-baboon renal xenografts. Transplantation 1998; 66 11):1439.

37. Xenotransplantation. Time to leave the laboratory. Lancet 1999;

2: 1657.38. Levy MF, Crippin J, Sutton S, et al. Liver allotransplantation

after extracorporeal hepatic support with transgenic (hCD55/hCD59) porcine livers: clinical results and lack of pig-to-hu-man transmission of the porcine endogenous retroviirus.Transplantation 2000; 69 (2): 272

39. Williams R. Classification, etiology and considerations of out-come in acute liver failure. Semin Liv Dis 1996; 16: 343.

40. Hammer GB, So SK, Al-Uzri A, et al. Continuous veno-venoushemofiltration with dialysis in combination with total hepatec-tomy and portocaval shunting. Bridge to liver transplantation.Transplantation 1996; 62 (1): 130.

Received 6 April 2000.Accepted 16 May 2000.

0041-1337/00/7007-998/0TRANSPLANTATION Vol. 70, 998–1005, No. 7, October 15, 2000Copyright © 2000 by Lippincott Williams & Wilkins, Inc. Printed in U.S.A.

MATRIX METALLOPROTEINASES CORRELATE WITHALVEOLAR-CAPILLARY PERMEABILITY ALTERATION IN

LUNG ISCHEMIA-REPERFUSION INJURY

PAOLA M. SOCCAL,1,7 YVAN GASCHE,2 JEAN-CLAUDE PACHE,3 ODILE SCHNEUWLY,4

DANIEL O. SLOSMAN,5 DENIS R. MOREL,4 ANASTASE SPILIOPOULOS,6 PETER M. SUTER,2 AND

LAURENT P. NICOD1

Division of Pulmonary Medicine, Department of Internal Medicine, Division of Surgical Intensive Care,Department of Anesthesiology, Pharmacology, and Surgical Intensive Care, Geneva University Hospital,

1211 Geneva 14, Switzerland

Background. Matrix metalloproteinases (MMPs) areable to degrade the endothelial basal lamina and in-crease vascular permeability.

Methods. In a porcine model of isolated-reperfusedlung, we studied the alveolar-capillary permeabilityand the zymographic expression of MMP-9 and MMP-2in the bronchoalveolar lavage fluid of lungs submittedex vivo to ischemia in three preservation solutions[modified Euro-Collins (EC), low-potassium-dextran,modified-blood]. Twenty-two pigs were randomly di-vided into three groups according to the preservationsolution used. One lung of each pig was rapidly reper-fused and analyzed (control lung) although the otherlung was reperfused and analyzed after 8 hr of isch-

emia (ischemic lung).Results. Alveolar-capillary permeability, evaluated

by the transferrin leak index, was increased after 8 hrof ischemia compared with controls in the threegroups, but was significantly higher in the modifiedEC group. In the EC group, after 8 hr of ischemia, bothproMMP-9 and MMP-9 increased significantly (8.8-and 22-fold, respectively) compared with controls andthis increase correlated with the transferrin leak in-dex. Neither proMMP-9 nor MMP-9 increased with theother two preservation solutions. The MMP-2 increaseafter ischemia was smaller and was also restricted tothe EC group.

Conclusion. MMP expression is enhanced duringlung ischemia-reperfusion, especially in the presenceof EC and this phenomenon correlates with the alter-ation of alveolar-capillary permeability.

During the past 15 years, lung transplantation has becomea well-established treatment for many advanced lung dis-eases. Ischemia-reperfusion (I-R) injury with graft dysfunc-tion and acute respiratory failure remains one of the mostfeared complications of the early postoperative period andthe primary cause of early morbidity and mortality (1). Al-though recent studies have suggested that most of the injury

1 Division of Pulmonary Medicine, Department of Internal Medi-cine

2 Division of Surgical Intensive Care.3 Divison of Clinical Pathology, Department of Pathology.4 Division of Anesthesiological Investigations, Department of An-

esthesiology, Pharmacology, and Surgical Intensive Care.5 Division of Nuclear Medicine, Department of Radiology.6 Unit of Thoracic Surgery, Department of Surgery.7 Address correspondence to: P. M. Soccal, MD, Division of Pulmo-

nary & Critical Care Medicine, Stanford University Medical Center,300, Pasteur Drive—H3143, Stanford, CA 94305-5236.

TRANSPLANTATION998 Vol. 70, No. 7