Antisense Oligodeoxynucleotides Prevent Acute Cardiac Allograft Rejection via a Novel, Nontoxic, Highly Efficient Transfection METHOD1,2

September 27, 1999

37.

38.

39.

40.

825

POSTON ET AL.

F. The immunosuppressive drug metabolite of leflunomide, A77 1726, affects murine T cells through two biochemical mechanisms. J Immunol 1997; 159: 22. Xu X, Williams JW, Gong H, Finnegan A, Chong AS- F. Two activities of the immunosuppressant, leflunomide: inhibition of pyrimidine synthesis and protein tyrosine phophorylation. Biochem Pharm 1996; 52: 527. Littler E, Stuart AD, Chee MS. Human cytomegalovirus UL97 open reading frame encodes a protein that phosphorylates the antiviral nucleoside analogue ganciclovir. Nature 1992; 358: 160. Sullivan V, Talarico CL, Stanat SC, Davis M, Coen DM, Biron KK. A protein kinase homologue controls phosphorylation of ganciclovir in human cytomegalovirus-infected cells. Nature 1992; 358: 162. Lalezari JP, Stagg RJ, Kuppermann BD, et al. Intravenous cidofovir for peripheral cytomegalovirus retinitis in patients

with AIDS. A randomized, controlled trial. Ann Intern Med 1997; 126: 257. 41. Grundy JE, Lawson KM, MacCormac LP, Fletcher JM, Youn KL. Cytomegalovirus- infected endothelial cells recruit neutrophils by the secretion of C-X-C chemokines and transmit virus by direct neutrophil-endothelial cell contact and during neutrophil transendothelial migration. J Infect Dis 1998; 177: 1465. 42. Rowe WP, Hartley JW, Waterman S, Turner HC, Huebner RJ. Cytopathogenic agent resembling human salivary gland virus recovered from tissue cultures of human adenoids. Proc Soc Exp Biol Med 1956; 92: 418. 43. Mladenovic V, Domljan Z, Rozman B, et al. Safety and effectiveness of leflunomide in the treatment of patients with active rheumatoid arthritis. Arthritis Rheumat 1995; 38: 1595. Received 21 December 1998. Accepted 25 March 1999.

0041-1337/99/6806-825/0

TRANSPLANTATION Copyright © 1999 by Lippincott Williams & Wilkins, Inc.

Vol. 68, 825– 832, No. 6, September 27, 1999 Printed in U.S.A.

ANTISENSE OLIGODEOXYNUCLEOTIDES PREVENT ACUTE CARDIAC ALLOGRAFT REJECTION VIA A NOVEL, NONTOXIC, HIGHLY EFFICIENT TRANSFECTION METHOD1,2 ROBERT S. POSTON,3,4 MICHAEL J. MANN,5 E. GRANT HOYT,4 MICHAEL ENNEN,4 VICTOR J. DZAU,5 AND ROBERT C. ROBBINS4,6 Department of Cardiothoracic Surgery, Stanford University School of Medicine, Stanford, California 94305, and Department of Cardiovascular Medicine, Brigham and Women’s Hospital, Harvard University Medical School, Boston, Massachusetts 02115

Background. We hypothesized that ex vivo donor allograft transfection with antisense oligodeoxynucleotide (AS ODN) would inhibit the expression of intercellular adhesion molecule (ICAM)-1, an important mediator of T-cell adhesion and costimulation, and therefore suppress acute cardiac rejection. Methods. Hearts were transfected ex vivo with AS, reverse AS ODN, or saline by applying 3 atm pressure for 45 min at 4°C. Grafts were then transplanted into allogenic recipients 6 treatment with leukocyte func-

1 This work was presented as a featured abstract at the International Society of Heart Lung Transplantation, London, England, March 1997. 2 This work was supported also in part by a grant from the Ralph and Mariam Falk Research Fund. 3 Fellow of the Thoracic Surgery Foundation for Research and Education. 4 Department of Cardiothoracic Surgery, Stanford University School of Medicine. 5 Department of Cardiovascular Medicine, Brigham and Women’s Hospital. 6 Address correspondence to: Robert C. Robbins, MD, Department of Cardiothoracic Surgery, Stanford University School of Medicine, Stanford, CA 94305. E-mail: [email protected].

tion-associated antigen (LFA)-1 monoclonal antibody (mAb) (1.5 mg/kg intravenously), cyclosporine (2.5 mg/ kg/day p.o.), or rapamycin (0.025 mg/kg/day intraperitoneally). Reperfusion injury was assessed in grafts harvested at early time points using the myeloperoxidase, %wet weight, and %contraction band necrosis assays; transfection efficiency was assessed using fluorescent microscopy; and efficacy of ICAM-1 blockade was assessed using immunohistochemistry. Other grafts were followed until rejection with donor/thirdparty skin grafting, adoptive transfer, and interleukin 2 infusion studies in selected recipients. Results. Transfection was highly efficient (fluorescein isothiocyanate-ODN in 4865% of total myocardial nuclei), nontoxic, and reduced the ICAM-1–positive area to 53614% versus having no effect on MHC class I expression (n54). The incidence of survival >60 days after AS ODN 1 LFA-1 monoclonal antibody was 75%, significantly higher than other regimens. Conclusion. AS ODN hyperbaric transfection proved highly efficient, effective at ICAM-1 blockade, and induced cardiac allograft tolerance when combined with LFA-1 monoclonal antibody. This highly targeted alteration of allograft immunogenicity may have an important role in future immunosuppressive strategies.

826

TRANSPLANTATION

Much of the morbidity and mortality that follows solid organ transplantation results from the use of relatively nonspecific immunosuppressive medications. Ex vivo gene therapy is a novel, highly targeted potential method of manipulating donor allograft immunogenicity. However, current methods of gene delivery using viral (1) or lipid (2) vectors have limitations such as poor transfection efficiency and unwanted toxicity that have prevented their widespread use in transplantation. Nondistending pressure has been used as a novel, practical, and nontoxic vector for nuclear delivery of phosphorothioate oligodeoxynucleotides (ODN*) to cardiovascular tissues (3– 6). This method has achieved a consistent and highly efficient delivery of DNA to cardiac allografts, and we therefore pursued an organ-specific immunosuppression regimen based on down-regulation of intercellular adhesion molecule (ICAM)-1. This particular cell surface receptor is an attractive immunosuppressive target because of the central role it is believed to play in allograft pathophysiology. An early increase in ICAM-1 expression after graft reperfusion initiates a feedback cytokine-adhesion molecule cascade that promotes nonspecific inflammatory processes such as reperfusion injury (7). In addition, the interaction of ICAM-1 with its ligand, leukocyte function-associated antigen (LFA)-1, is an important pathway of T-cell costimulation (8). Therefore, ICAM-1 influences alloantigen-specific processes such acute rejection and potentially chronic rejection as well (5, 9, 10). In experimental cardiac allograft models, blockade of the ICAM1/LFA-1 pathway by the systemic administration of LFA-1 and ICAM-1 monoclonal antibodies (mAbs) leads to reduced reperfusion injury (11) and donor-specific allograft tolerance (12). The expression of ICAM-1 is transcriptionally regulated and extremely low in myocardial tissue at baseline before transplantation (8), features that make ICAM-1 highly amenable to an antisense (AS) strategy. Previously, AS ICAM-1 ODN has been delivered by systemic infusion to cardiac allograft recipients to block ICAM-1 up-regulation and prolong graft survival (13). By avoiding the need for systemic ICAM-1 blockade, ex vivo donor graft transfection with AS ODN decreases side effects such as infectious risk to the recipient, while providing an equally potent form of immunosuppression. In this study, donor hearts were pressure transfected with AS ODN against ICAM-1 to assess the feasibility and biologic efficacy of this gene therapy approach. The immunologic effects of this therapy were further analyzed using the PVG-to-ACI rodent cardiac model. MATERIALS AND METHODS Drugs and agents. Phosphorothioate ODN or fluorescein isothiocyanate (FITC)-labeled phosphorothioate ODN (Biosource, Menlo Park, CA) were used at a final concentration of 80 mM. We chose a sequence for the AS ODN used in these studies that was specific for the inhibition of rat ICAM-1 mRNA by modifying the previously * Abbreviations: AS, antisense; BN, Brown Norway; CBN, contraction band necrosis; CsA, cyclosporine; DS, donor-sensitized splenocytes; FITC, fluorescein isothiocyanate; HrIL, human recombinant interleukin; ODN, oligodeoxynucleotide; ICAM, intercellular adhesion molecule; i.p., intraperitoneal(ly); i.v., intravenously; LFA, leukocyte function-associated antigen; mAb, monoclonal antibody; MPO, myeloperoxidase; MST, mean survival time; POD, postoperative day; RAS, reverse antisense.

Vol. 68, No. 6

reported mouse AS ICAM-1 ODN (IP3082). Using the previously published sequences of rat and mouse ICAM-1 mRNA, the AS sequence was altered to reflect the following differences at this homologous 39-untranslated target region: Mouse ICAM-1 mRNA (position 11755–74): 59ATG GTG GCC TGG GGG ATG CA 39 AS mRNA (IP3082): 39TAC CAC CGG ACC CCC TAC GT 59 Rat ICAM-1 mRNA (position 11768 – 87): 59GACA GTG GCC TGG GGA TGC A 39 AS mRNA (used in our study): 39CTGT CAC CGG ACC CCT ACG T 59 (bold type represents bases that differ between mouse and rat ICAM-1 mRNA and AS at this region). Scrambled and reverse antisense (RAS) sequences were used as controls. LFA-1 and ICAM-1 mAbs were purified from the ascites of mice injected with hybridomas (WT.1 and 1A29; gift of M. Miyasaka, Osaka, Japan) and administered via the dorsal penile vein at 1.5 mg/kg/day from postoperative day (POD) 0 – 6. This dose was confirmed to promote full saturation of their respective ligands on peripheral blood leukocytes 24 hours after administration by flow cytometric analysis (n53). Cyclosporine (CsA; Sandoz, East Hanover, NJ) was dissolved in olive oil (10 mg/ml) and administered via gavage at a dose of 5 mg/kg for 10 days postoperatively to prevent acute rejection and allow for prolonged graft survival. Rapamycin (gift of S.N. Sehgal, Wyeth-Ayerst) was supplied in pure form and stored at 4°C. It was suspended in 0.2% carboxymethyl-cellulose and administered at a dose of 0.025 mg/kg/day for 7 days via intraperitoneal (i.p.) injection. Human recombinant interleukin (HrIL)-2 (gift of Chiron Corp., Emeryville, CA) was supplied as a lyophilized product and reconstituted with sterile water. It was administered on either POD 5 or POD 60 at a dose of 10,000 IU t.i.d. 3 5 days to test for T-cell anergy as previously described (14). Animals. Adult male (8 to 10 weeks old, 230 –270 g) PVG (RT1c) and ACI (RT1a) rats were obtained from Harlan Sprague Dawley (Indianapolis, IN) and maintained at 21° 6 2°C in the animal care facilities of Stanford University Medical Center (Stanford, CA). All animals received humane care in compliance with the Principles of Laboratory Animal Care, formulated by the National Society for Medical Research, and the Guide for the Care and Use of Laboratory Animals, prepared by the National Academy of Sciences and published by the National Institutes of Health (NIH Publications no. 86-23, revised 1985). Heart transplantation. Both donor and recipient rats were anesthetized with methoxyflurane (inhalational) and sodium pentobarbital (50 mg/kg i.p.). Donor rats were then heparinized (50 mg/kg intravenously [i.v.]), and the hearts were procured and transfected as described below. Grafts were heterotopically placed into the abdomen of allogenic recipients using a modification of the methods described by Ono and Lindsay (15). Immediately after reperfusion and daily thereafter, grafts were assessed for successful return of rhythmic cardiac contractions on a scale ranging from 0 (no contractions) to 4 (vigorous contractions) in a manner blinded to treatment regimen. Primary failure was defined in grafts that did not achieve a palpation score .2 within 1 hr. After an initial period of successful function, hearts were considered acutely rejected when palpation scores were ,1. Grafts with either of these diagnoses were excluded from further study. In vivo transfection efficiency. After donor rat median sternotomy, an aortic cross-clamp was applied and 2 ml of ice-cold Stanford cardioplegia was infused proximal to the clamp for coronary perfusion. After complete cardioplegic arrest, FITC-tagged scrambled ODN was infused (0.5 ml of 80 mM ODN in phosphate-buffered saline solution at 1 ml/min) proximal to the aortic cross-clamp for coronary perfusion. The heart was then explanted, placed into a well of ODN solution in an ice bath at 4°C inside a custom-designed vessel. The vessel was pressurized at a rate of 2 atm/minute to a peak of 3 atm above ambient pressure. After 45-min incubation, the vessel was depressurized at a rate of 1 atm/minute, and the grafts were hetero-

September 27, 1999

827

POSTON ET AL.

topically transplanted. The optimization of the transfection variables (dose of pressure and ODN, incubation time, etc.) has been reported previously (14). After varying reperfusion times (6 hr to 7 days), grafts with palpation scores .2 were procured, flushed with ice-cold phosphate-buffered saline, and divided longitudinally. Tissue samples were immediately snap-frozen in OCT embedding compound in liquid nitrogen. After equilibration to 220°C, 6-mm sections were taken perpendicular to the long axis of the heart from the superior, midportion, and inferior portions of the grafts. Using Hoechst dye to facilitate the identification of total nuclei under fluorescent microscopy, transfection efficiency was assessed by determining the percentage of myocardial nuclear localization of fluorescence in five random high-powered (3100) fields chosen systematically from each of three graft regions (superior, midportion, and inferior portions). Method toxicity. Donor grafts that underwent the full transfection method (1ischemia, 1pressure) were compared with grafts treated with the above protocol without the use of pressure (1ischemia, 2pressure) or by immediate transplantation without a pressure incubation period (2ischemia, 2pressure). To assess the toxicity of the hyperbaric method, groups were analyzed for evidence of immediate and delayed graft injury. Immediate injury was defined as a palpation score ,2 at 1 hr after release of aortic cross-clamp. Delayed injury at 6 and 12 hr after reperfusion was assessed by three parameters: (1) cardiac edema using percent wet weight assay (%w/ w) 2 heart weighed before and after drying in oven at 100°C for 24 hr, (2) neutrophil infiltration assessed by myeloperoxidase (MPO) enzyme activity 2 total protein isolated from frozen tissue and assayed for change in absorbance at 470 nm at 1 min after addition of guaiacol and peroxide, and (3) histologic injury as determined by the percentage involvement with contraction band necrosis (%CBN)—after initial review with a pathologist, the percentage cross-sectional area of a trichrome-stained section involved with CBN was quantitatively determined using computer-assisted image analysis (C-imaging systems, Cranberry Township, NJ). Verification of transfection efficacy. PVG grafts were transfected with AS ICAM-1 ODN or RAS ODN and heterotopically transplanted into the abdomen of untreated ACI recipients. On the basis of pilot studies of ICAM-1 protein expression analyzed in control allografts by Western blot and immunohistochemistry, POD 3 was chosen as the time point most appropriate for comparison with AS ODN-transfected allografts. Hearts were procured on POD 3 for immunohistochemistry assessment of ICAM-1, MHC class I and II using the method outlined in the Histostain SP kit (Zymed Laboratories, South San Francisco, CA). Briefly, 6-mm sections were air dried at room temperature and fixed in acetone at 220°C. Sections were rehydrated in 1% bovine albumin/phosphate-buffered saline and then incubated with one of following primary antibodies (from Serotec, Westbury, NY): 1A29 (ICAM-1), 156 and 280 (MHC class I), and 46 (MHC class II). This was followed by incubation with a biotinylated goat anti-mouse IgG (Zymed Laboratories). The avidin-biotin complex was applied, and diaminobenzidine tetrahydrochloride was used as the chromogen. The substitution of 1% bovine albumin/phosphatebuffered saline for the primary antibody served as the negative (reagent) control. Rat cervical lymph nodes served as the ICAM-1– positive control. Sections from transfected and control grafts were scored (0 to 31) for ICAM-1, MHC class I and II staining intensity by a pathologist blinded as to experimental group. Computer-assisted image analysis (C-imaging systems) was used to provide a quantitative measurement of the AS effect by measuring the total graft cross-sectional area positive for ICAM-1. Biologic effects of ex vivo AS ICAM-1 ODN. The biologic effect of ex vivo AS ICAM-1 ODN transfection on acute rejection was assessed by daily abdominal palpation of heterotopically placed cardiac allografts. Grafts were treated with either AS ODN, RAS ODN, or saline. Recipients were either untreated or treated with LFA-1 mAb 1.5 mg/kg i.v. 3 7 days, CsA 2.5 mg/kg/day 3 10 days, or rapamycin 0.025 mg/kg/day i.p. 3 7 days. For an additional positive control, a group was treated with a combination of ICAM-1 mAb and LFA-1

mAb (1.5 mg/kg i.v. 3 7 days). Groups were assessed for mean survival time (MST) and the presence of prolonged graft acceptance, which was defined as a vigorously beating graft (palpation score $3) at .60 days postoperatively. Mechanism of action of AS ICAM-1 ODN and LFA-1 mAb. The ex vivo AS ODN and LFA-1 mAb group (AS/mAb group, n520) was compared with a group treated with CsA 10 mg/kg p.o. on days 0 –9 (CsA-positive control group), which has been shown to induce allograft-specific, suppressor cell-mediated tolerance in this rat strain combination (16). To assess the specificity of the prolonged allograft acceptance, recipients were secondarily grafted on POD .60 with skin from donor (PVG) and third-party (Brown Norway; BN) strain rats on both flanks. For negative control, a group of naive ACI rats were grafted with PVG and BN skin, which also served as a source of donor-sensitized splenocytes (DS) after the untreated rejection of their skin grafts. Acute skin graft rejection was judged postoperatively according to previously described criteria (16). The intraperitoneal administration of HrIL-2 (Chiron Corp.) has been shown to reverse T-cell anergy established in rodents in vivo (17), one possible mechanism of prolonged acceptance in this model. Mean allograft survival time in the AS/mAb group was assessed after injection with HrIL-2 (10,000 IU i.p. t.i.d.) on POD 5–10 (n56) or POD 60 – 65 (n510) by daily abdominal palpation. In addition, adoptive transfer techniques were used to test for the presence of suppressor cells as described (16). Splenocytes were harvested from the CsA-positive control, AS/mAb, and DS groups using previously described techniques. These cells were then infused into ACI rats tolerized to PVG cardiac grafts and then irradiated (total lymphocyte irradiation protocol) at a dose of 200 rad 3 5 days. The following combinations of splenocytes were used: (1) 103106 DS alone (negative control, n56); (2) 503106 CsA-positive control 1 103106 DS (positive control, n56); and (3) 503106 AS/mAb 1 103106 DS (experimental group, n56). Tolerance reversal (i.e., PVG cardiac allograft rejection) was assessed daily after splenocyte infusion for each group by abdominal palpation. Statistical analysis. Nonparametric data (mean graft survival time and mean functional palpation score at 30 min) was compared between groups using the Mann-Whitney U test with further comparison of the incidences of prolonged graft acceptance (graft survival .60 days after treatment) and primary graft failure (palpation score ,2 at 30 min) using the Fisher Exact Test. MPO levels, %w/w, and %nuclear fluorescence were analyzed using repeated measures analysis of variance with a post-hoc t test. Given the consistently unequal SD noted in the mean areas of %CBN and %ICAM-1 staining between the AS-transfected and untransfected groups, the Welch t-test was required for comparison of these data. Statistical significance was assigned to two-tailed p values less than 0.05. RESULTS

Efficiency of transfection. After pressure-mediated transfection of FITC-ODN, nuclear localization of fluorescence was detected in 4865% of total myocardial cells after 24 hr of reperfusion (n55) (Table 1). However, transfection appeared to be nonuniform, with some areas demonstrating very little efficiency, while others (such as that shown in the representative high-powered field of Fig. 1) showing dense nuclear fluorescence in close to 90% of total cells. Less total nuclear fluorescence was seen after longer periods of reperfusion, however, persistence of fluorescence was noted up to 7 days (5.364.0%, n53). A 45-min incubation with FITC-ODN solution without pressure (1ischemia, 2pressure) did not achieve detectable transfection (0%, n55). Toxicity. The full hyperbaric transfection method (1ischemia, 1pressure) did not provoke an increase in immediate graft injury compared with control groups undergoing the identical protocol but without pressure (1ischemia, 2pres-

828

TRANSPLANTATION

Vol. 68, No. 6

FIGURE 1. Hyperbaric transfection led to highly efficient myocyte delivery of ODN as illustrated by the examination of each random high-powered field (3100) after Hoechst dye staining to demonstrate total myocardial nuclei (A) and under the fluorescent filter (B, 3100) to demonstrate nuclear localization of fluorescence. Intracoronary infusion of FITC-ODN without application of pressure did not lead to appreciable nuclear localization of fluorescence (data not shown).

sure) or ischemia (2ischemia, 2pressure). After 30-min reperfusion, the incidence of primary graft failure and functional palpation score following the full method (8.3% and 3.460.9, respectively, n540) was not significantly different from that seen by excluding pressure (10.0%, 3.361.1, n520) or both pressure and ischemia (7.5%, 3.261.2, n512). Furthermore, delayed graft injury was also similar between groups. At 6 or 12 hr, no evidence of increased delayed myocardial injury in terms of neutrophil infiltration (MPO), cardiac edema (%w/w), or histologic injury (%CBN) was seen following the full method compared with the other groups (Table 2). Efficacy of transfection. At day 3, the AS ODN treatment resulted in an effective but nonuniform blockade of ICAM-1 up-regulation compared with control grafts (Fig. 2). Untransfected PVG grafts displayed diffuse up-regulation of ICAM-1 on immunohistochemical analysis (shown by the dark red staining) after 3 days in untreated ACI recipients. Using image analysis, ex vivo AS transfection was found to significantly (although inhomogeneously) reduce ICAM-1–positive staining to 53614% of the total cross-sectional area. This percentage reduction was similar to the percentage of myocardium that had been found to be successfully transfected with FITC-ODN. The expression of MHC class I and II was

unaffected in AS ODN-treated grafts compared with control grafts (data not shown). Biologic effect of ICAM-1 blockade by AS ODN. The group of rats receiving AS ODN transfection alone did not demonstrate prolonged graft acceptance (.60 days) or a significant increase in MST compared with the untreated control group (7.861.2 days, n56 vs. 8.862.6 days, n56). However, the combination of ex vivo AS ODN therapy with recipient LFA-1 mAb treatment induced a significantly higher incidence of prolonged acceptance (75%, n527) than any other group (Table 3). The MST of AS ODN-transfected grafts was significantly longer after combination with recipient LFA-1 mAb (.85.7635.1 days) compared with CsA (12.761.5 days, n56) or rapamycin (12.061.8 days, n56). The lack of prolonged acceptance after the substitution of RAS ODN in the LFA-1 mAb-treated recipients supports a true AS mechanism of action for the AS/mAb group. Blockade of ICAM-1 using systemic mAb instead of ex vivo AS ODN in LFA-1 mAbtreated recipients also increased the incidence of prolonged acceptance (50%, n510) and MST (.60.5643.7 days) (Table 3). Mechanism of AS ICAM/LFA-1 mAb effect. The donor specificity of prolonged acceptance was confirmed by secondarily placed donor (PVG) and third-party (BN) skin allo-

TABLE 1. Transfection efficiency Reperfusion time

n

Number transfected nuclei (Number fluorescent staining/graft)

Number total nuclei (Number Hoechst staining/graft)

% Fluorescence (Number fluorescent/total)

24 hr 3 days 7 days

5 4 3

30056487 19056579 3836292

62326378 70046243 72286390

48.5 27.2 5.3

Allografts were transfected with fluorescently labeled oligos (FITC-ODN), heterotopically transplanted and then removed after 1, 3, and 7 days of reperfusion. Frozen sections were stained with Hoechst dye to identify the number of total nuclei and analyzed for transfection efficiency using fluorescent microscopy.

September 27, 1999

829

POSTON ET AL.

FIGURE 2. On POD 3, immunohistochemistry demonstrated a significant but inhomogeneous blockade of ICAM-1 expression in AS ODNtransfected allografts (A) compared with the diffuse expression of ICAM-1 throughout the graft after RAS ODN transfection (B) (320). TABLE 2. Toxicity studies of hyperbaric transfection Reperfusion times (hr) Ex vivo conditions

MPO 2Ischemia, 1Ischemia, 1Ischemia, %w/w 2Ischemia, 1Ischemia, 1Ischemia, %CBN 2Ischemia, 1Ischemia, 1Ischemia,

Baseline (n53)

16° (n54)

112° (n55)

2pressure 2pressure 1pressure

0.00860.020 0.00860.011 0.01060.014

0.1060.06 0.1260.06 0.1060.11

0.0960.04 0.0760.04 0.0960.03

2pressure 2pressure 1pressure

68.3861.29 69.7860.93 70.2361.23

79.2162.04 80.0260.82 79.7761.32 79.2461.21 81.0261.89 80.9160.93

2pressure 2pressure 1pressure

0 0 0

8.9562.16 10.5461.86 10.3061.15 12.0062.62 10.0261.69 11.7562.32

Grafts exposed to the transfection protocol (1ischemia, 1pressure) were procured just before heterotopic transplantation (baseline) and after 6 and 12 hr of reperfusion to assess for three toxicity parameters: MPO activity (U/mg), %w/w, and %CBN. No significant differences were seen compared with the control groups (6ischemia, 2pressure; analysis of variance).

grafts. The AS/mAb group displayed significantly delayed kinetics of PVG donor skin rejection compared with that seen in the untreated, negative control group (29.562.8 days, n56 vs. 8.061.0, n56, P,0.01, Mann-Whitney U test), a pattern that was similar to that seen in CsA-positive controls (Table 4). In contrast, the AS/mAb group maintained the ability to reject third-party BN skin similar to CSA-positive control (11.262.0 days, n56 and 8.360.6 days, respectively). The use of IL-2 in the AS/mAb group did not significantly reverse tolerance (POD 60 – 65 treatment, n510) nor prevent it from being developed (POD 5–10 treatment, n56) (Table 5). On the other hand, cells harvested from AS/mAb group recipients were found, similar to the CsA-positive control group, to actively suppress rejection when adoptively transferred. The reversal of PVG allograft tolerance after DS splenocyte infusion alone (MST 9.162.5 days, n56) was prevented by the coinfusion of either AS/mAb (MST .86.5633.1 days, n56) or CsA (MST .100 days, n56) splenocytes with DS cells (Table 6). The close similarities between the CsA-positive control and AS/mAb groups in all these studies support a mechanism

for AS ODN 1 LFA-1 mAb therapy that includes the induction of active suppressor cells rather than T-cell anergy. DISCUSSION

Somatic gene therapy has recently shown promise in the rat transplantation model as a novel method of highly targeted immunosuppression (18). Solid organ transplantation is well suited for the use of this type of gene therapy. The organ procurement process allows for ex vivo DNA delivery that has the potential to dramatically improve transfection efficiency while avoiding undesirable systemic transfection. The transient nature of gene therapy seen in other models may actually be an advantage in transplantation, where some of the critical immunologic reactions to be targeted are consistently focused on the early postoperative period. The studies described in this report explored the utility of increased pressure as a highly efficient ex vivo transfection vector in which almost 50% of myocardial cells demonstrate nuclear localization of ODN. Viral and lipid vectors, on the other hand, have often shown low efficiency in myocardial tissues with transfection of less than 10% of total cells (1, 2). Despite this low efficiency, biologic effects on allograft rejection have been achieved using these vectors by overexpression of local immunosuppressive factors such as transforming growth factor-b or IL-10 (1). Unlike plasmids or viruses, transfection with AS oligos does not result in the paracrine release of locally protective factors. Therefore, the biologic success of an AS strategy mandates high transfection efficiency. In contrast to the potential toxicity (immunologic and chemical) of standard DNA delivery vectors (19), our studies found that pressurization of cardiac tissue did not result in measurable immediate or delayed myocardial damage. The benign nature of pressure-mediated transfection at low-intensity hyperbarics ,10 atm is consistent with whole animal (20) and human (21) hyperbaric studies. The mechanism by which pressure facilitates nuclear uptake of ODN is unknown but likely involves an alteration in cellular membrane permeability. We are currently investigating the speculation that this alteration is a change in the physical state of some membrane constituent(s) or a temporary sonication of the membrane during depressurization of tissue. Regardless, the ease of application of this novel vector to donor allografts combined with high efficiency and lack of overt toxicity pro-

830

TRANSPLANTATION

Vol. 68, No. 6

TABLE 3. Combination of ICAM-1 AS ex vivo with recipient LFA-1 mAb leads to prolonged allograft survival Ex vivo treatment

Postoperative treatment

n

Saline AS ICAM-1 Saline

None None LFA-1 mAb

6 6 12

AS ICAM-1

LFA-1 mAb

27

RAS ICAM-1 Saline Saline AS ICAM-1 Saline AS ICAM-1

LFA-1 mAb LFA-1 1 ICAM mAb CsA CsA Rapamycin Rapamycin

6 6 6 6 6 6

Survival (days)

Tolerance (% of grafts)

MST (days)

0 0 17

8.862.6 7.861.2 27.2634.0

75a

.85.7635.1b

0 50a 0 0 0 0

12.361.5 .60.5643.7b 12.262.6 12.761.5 11.762.6 12.061.8

6, 7(32), 9, 12 (32) 7 (33), 8 (32), 10 12 (35), 13 (34), 14, .100 (32) 12 (33), 13 (33), 14, .100 (320) 11 (32), 12 (32), 13, 15 12, [19], [32], .100 (33) 8, 11 (32), 14 (32), 15 11 (32), 13 (33), 15 8, 9, 12, 13 (32), 15 9, 11, 12, 13 (32), 14

PVG allografts were transfected using saline (control), antisense oligo (AS ICAM-1), or reverse antisense oligo (RAS ICAM-1) and heterotopically grafted into recipients that were untreated or treated with a short course of LFA-1 mAb (1.5 mg/kg), CsA (2.5 mg/kg), or rapamycin (0.025 mg/kg). Graft survival and MST were assessed by daily abdominal palpation. Tolerance was defined by survival at .100 days. Brackets around number indicates recipient died with beating graft. a P,0.01 vs. all other groups, two-tailed Fisher test. b P,0.03 vs. all other groups, two-tailed Mann Whitney U test. TABLE 4. Delayed rejection of donor type compared with thirdparty skin grafts in tolerant recipients MST (days)

Groupa N

Third-party skin (BN)

Donor skin (PVG)

11.262.0 8.360.6 8.261.3

29.562.8† 23.363.1† 8.061.0

AS/mAb therapyb 6 CsA-positive controlc 6 Negative controld 6

a Skin grafts from third-party or donor strain rats were placed on each flank of ACI rats previously made tolerant to PVG hearts following b ex vivo AS ODN1LFA-1 mAb (AS/mAb) or c CSA 10 mg/kg p.o.310 days (positive control). d Naive ACI rats served as the negative control skin graft recipients. e P,0.05 versus negative control donor skin, Mann-Whitney U test.

TABLE 6. Adoptive transfer of tolerance after ICAM-1/LFA-1 blockade Type of splenocytes infused

n

DSa alone (103106) DS (103106)1CSA-positive controlb (503106) DS (103106)1AS/mAbc (503106)

6 6

9.261.6 (7, 934, 12) .100 (6)

MST (days after infusion)†

6

.86.5633.1 (19, .10035)

Recipients for the adoptive transfer studies were ACI rats tolerized to PVG grafts by short course CsA that were irradiated (TLI 200 rad35 days) on POD .100. Splenocytes for infusion were harvested from ACI rats exposed to PVG alloantigen under the following conditions: a no treatment (donor skin sensitized, DS), bshort course CsA (positive control), or c ex vivo AS ODN and LFA-1 mAb (AS/mAb). After infusion, the heterotopic grafts of the adoptive transfer recipients were palpated daily.

TABLE 5. Absence of T-cell anergy IL-2 treatment date

Group

n

CsA-positive controla AS/mAbb AS/mAbb

10

POD 5–10

10 6

POD 60–65 POD 5–10

MST (days after IL-2 Rx)

.100 (310) .100 (310) .67.7650.1 (2, 4, .10034)

ACI recipients that were treated with a short course CsA (positive control) or bex vivo AS ODN/LFA-1 mAb (AS/mAb) were retreated with IL-2 (10,000 IU i.p. t.i.d.36 days). Survival of the PVG heart graft was not significantly affected by IL-2 administered either in the early (POD 5–10) or late (POD 60 – 65) period after transplantation.

vides a clear technologic advantage over existing vectors for use in transplantation. Although ODN transfection has been associated with nonsequence-specific effects on cellular proliferation (22, 23), which could potentially influence allograft outcome, a true AS mechanism on ICAM-1 expression in the studies reported here was indicated by: (1) the lack of effect of control RAS ODN on either target protein expression or acute rejection and (2) the lack of nonspecific blockade of other gene products also involved in allograft rejection (e.g., MHC class I and II). In addition, an alternative method of ICAM-1 blockade that

involved systemic exposure to an mAb confirmed that inhibition of both ICAM-1 and LFA-1 could yield a similar incidence of prolonged graft acceptance in this model. Verification of an effective and sequence-specific method of ex vivo ICAM-1 blockade provided a rational basis for exploring the effect of our ICAM-1 AS therapy on cardiac transplantation. A number of studies have probed the role of ICAM-1 in mediating cardiac transplant rejection. Allograft survival was not prolonged in ICAM-1 knockout mice (24), despite the success of ICAM-1 mAb or systemic ICAM-1 AS ODN monotherapy in delaying rejection in other studies involving this species (12, 13). Ex vivo ICAM-1 AS ODN monotherapy did not influence graft survival in this study in rats, a finding consistent with other rat anti-ICAM-1 monotherapies (25, 26). In all of these studies, however, the addition of LFA-1 mAb to anti-ICAM-1 therapy has provided greater immunosuppressive synergy and stronger induction of prolonged graft acceptance than any other agent (e.g., CsA, rapamycin). In fact, the synergy of this immunosuppressive strategy used in our study was seen after only 50% ICAM-1 blockade in the grafts of LFA-1 mAb-treated recipients. The simultaneous blockade of both axes of this form of immune cell interaction with donor myocytes may therefore provide

September 27, 1999

POSTON ET AL.

for an effective means of altering transplant biology long term. This model involves transplantation across a full MHC mismatch (i.e., RT1c to RT1a). However, results of this study may have been influenced by the well-described “low responsiveness” of the ACI recipient rat. For instance, MST of the positive control group (ICAM-1 1 LFA-1 mAb treated) was disproportionately prolonged compared with models using other rat strain combinations (27, 28). In addition, the IL-2 resistance of PVG allografts in the AS/mAb group is inconsistent with T-cell anergy, the mechanism of effect for blockade of this costimulatory pathway that has been demonstrated in in vitro studies (29). These ACI recipients were found to have donor-specific, tolerizing splenocytes that could be adoptively transferred, an immune response alteration most consistent with an active suppressor cell hypothesis and true allograft tolerance induction. Of interest, anti-rhesus LFA-1 mAb monotherapy prolongs cardiac allograft survival in rhesus monkeys as effectively as in this rodent model (23.062.6 days vs. 27.2634.0 days, monkey vs. rat, respectively) (Poston, unpublished observations, 1998). Taken together, the results of this study and the effectiveness of the LFA-1 mAb in rhesus monkeys support ongoing investigations of ex vivo AS ICAM-1 ODN and LFA-1 mAb in the more clinically relevant nonhuman primate model. Nonetheless, optimization of ICAM-1 blockade by further improving transfection, using alternate AS sequences (30), and further investigation in strong responder rodents is warranted. In conclusion, these studies outline the development and demonstrate the feasibility of this simple, reproducible, efficient, and nontoxic ODN delivery system. When combined with LFA-1 mAb, use of ex vivo AS ICAM-1 ODN instead of systemic ICAM-1 mAb was found to provide a more graftspecific, yet equally potent form of immunosuppression. We feel that this highly targeted method of altering donor organ immunogenicity may reduce the need for recipient nonspecific immunosuppression and play an important future role in clinical immunosuppressive strategies.

REFERENCES 1. Qin L, Chavin KD, Ding Y, et al. Multiple vectors effectively achieve gene transfer in a murine cardiac transplantation model: immunosuppression with TGF-beta 1 or vIL-10. Transplantation 1995; 59: 809. 2. Dalesandro J, Akimoto H, Gorman CM, et al. Gene therapy for donor hearts: ex vivo liposome-mediated transfection. J Thorac Cardiovasc Surg 1996; 111: 416; discussion 421. 3. Poston R, Tran K, Mann M, Hoyt E, Dzau V, Robbins R. Use of ex vivo gene therapy to block ischemically induced neointimal hyperplasia. J Heart Lung Transplant 1998; 17: 349. 4. Mann MJ, Kernoff R, Dzau VJ. Vein graft gene therapy using E2F decoy oligonucleotides: target gene inhibition in human veins and long term resistance to atherosclerosis in rabbits. Surg Forum 1997; 48: 274. 5. Poston R, Ennen M, Pollard J, Hoyt E, Billingham M, Robbins R. Ex vivo gene therapy prevents chronic graft vascular disease in cardiac allografts. J Thorac Cardiovasc Surg 1998; 116: 128. 6. Pollman MJ, Hall JL, Mann MJ, Zhang L, Gibbons GH. Inhibition of neointimal cell bcl-x expression induces apoptosis and regression of vascular disease. Nat Med 1998; 4: 222. 7. Poston R, Ing D, Pollard J, Hoyt E, Robbins R. Upregulation of

8. 9.

10.

11.

12.

13.

14.

15. 16.

17.

18.

19.

20.

21.

22.

23. 24.

25.

26.

27.

831

ICAM-1 leads to increased reperfusion injury after global cardiac ischemia. Surg Forum 1996; 47: 233. Springer TA. Adhesion receptors of the immune system. Nature 1990; 346: 425. Poston R, Billingham M, Pollard J, Hoyt E, Robbins R. Effect of ICAM-1 on reperfusion injury and chronic graft vascular disease. Ann Thorac Surg 1997; 64: 1004. Land W, Messmer K. The impact of ischemia/reperfusion injury on specific and non-specific, early and late chronic events after organ transplantation. Transplant Rev 1996; 10: 108. Tamiya Y, Yamamoto N, Uede T. Protective effect of monoclonal antibodies against LFA-1 and ICAM-1 on myocardial reperfusion injury following global ischemia in rat hearts. Immunopharmacology 1995; 29: 53. Isobe M, Yagita H, Okumura K, Ihara A. Specific acceptance of cardiac allograft after treatment with antibodies to ICAM-1 and LFA-1. Science 1992; 255: 1125. Stepkowski SM, Tu Y, Condon TP, Bennett CF. Blocking of heart allograft rejection by intercellular adhesion molecule-1 antisense oligonucleotides alone or in combination with other immunosuppressive modalities. J Immunol 1994; 153: 5336. Feeley BT, Vriens PW, Hoyt EG, Ennen MP, Robbins RC. Increased pressure for ex vivo transfection of antisense oligodeoxynucleotides in cardiac allografts improves transfection efficiency. Transplantation 1998; 65 (suppl): S182. Ono K, Lindsey ES. Improved technique of heart transplantation in rats. J Thorac Cardiovasc Surg 1969; 57: 225. Nisco SJ, Hissink RJ, Vriens PW, Hoyt EG, Reitz BA, Clayberger C. In vivo studies of the maintenance of peripheral transplant tolerance after cyclosporine: radiosensitive antigen-specific suppressor cells mediate lasting graft protection against primed effector cells. Transplantation 1995; 59: 1444. Dallman MJ, Wood KJ, Morris PJ. Recombinant interleukin-2 (IL-2) can reverse the blood transfusion effect. Transplant Proc 1989; 21(1 Pt 1): 1165. Li XK, Okuyama T, Tamura A, et al. Prolonged survival of recipient rats with Fas-ligand-transfected liver allografts by using HVJ-liposome. Transplant Proc 1998; 30: 943. Schneider DB, Fly CA, Dichek DA, Geary RL. Adenoviral gene transfer in arteries of hypercholesterolemic nonhuman primates. Hum Gene Ther 1998; 9: 815. Chavko M, Harabin A. Regional lipid peroxidation and protein oxidation in rat brain after hyperbaric oxygen exposure. Free Radic Biol Med 1996; 20: 973. Harabin A, Survanshi S, Homer L. A test for variations in individual sensitivity to hyperbaric oxygen toxicity. Undersea Hyberb Med 1994; 21: 403. Guvakova MA, Yakubov LA, Vlodavsky I, Tonkinson JL, Stein CA. Phosphorothioate oligodeoxynucleotides bind to basic fibroblast growth factor, inhibit its binding to cell surface receptors, and remove it from low affinity binding sites on extracellular matrix. J Biol Chem 1995; 270: 2620. Stein CA, Cheng YC. Antisense oligonucleotides as therapeutic agents: is the bullet really magical? Science 1993; 261: 1004. Schowengert K, Zhu J, Stepkowski S, Tu Y, Entman M, Ballantyne C. Cardiac allograft survival in mice deficient in intercellular adhesion molecule-1. Circulation 1995; 92: 82. Stepkowski SM, Wang ME, Amante A, et al. Antisense ICAM-1 oligonucleotides block allograft rejection in rats. Transplant Proc 1997; 29(1–2): 1285. Takazawa K, Hosoda Y, Bashuda H, et al. Synergistic effects of mycophenolate mofetil (MMF, RS-61443) and anti-LFA-1/ ICAM-1 monoclonal antibodies on the prolongation of heart allograft survival in rats. Transplant Proc 1996; 28: 1980. Brandt M, Steinmann J, Steinhoff G, Haverich A. Treatment with monoclonal antibodies to ICAM-1 and LFA-1 in rat heart allograft rejection. Transpl Int 1997; 10: 141.

832

TRANSPLANTATION

28. Tsuchida Y, Usui M, Naitoh T, Takahashi T, Murakami M, Uede T. Limb allografts in rats treated with anti-ICAM-1 and antiLFA-1 monoclonal antibodies. J Reconstr Microsurg 1997; 13: 107. 29. Isobe M, Suzuki J, Yamazaki S, Sekiguchi M. Effects of exogenous IL-2 on tolerance induction by anti-ICAM-1 and antiLFA-1 monoclonal antibodies. Transplant Proc 1996; 28: 1045.

Vol. 68, No. 6

30. Stepkowski SM, Wang ME, Condon TP, et al. Protection against allograft rejection with intercellular adhesion molecule-1 antisense oligodeoxynucleotides. Transplantation 1998; 66: 699.

Received 26 January 1999. Accepted 21 April 1999.

0041-1337/99/6806-832/0

TRANSPLANTATION Copyright © 1999 by Lippincott Williams & Wilkins, Inc.

Vol. 68, 832– 839, No. 6, September 27, 1999 Printed in U.S.A.

REGULATED INHIBITION OF COAGULATION BY PORCINE ENDOTHELIAL CELLS EXPRESSING P-SELECTIN-TAGGED HIRUDIN AND TISSUE FACTOR PATHWAY INHIBITOR FUSION PROTEINS DAXIN CHEN,1 KRISTIAN RIESBECK,1–3 JOHN H. MCVEY,2 GEOFFREY KEMBALL-COOK,2 EDWARD G. D. TUDDENHAM,2 ROBERT I. LECHLER,1 AND ANTHONY DORLING1, 4 Department of Immunology and MRC Clinical Sciences Centre Haemostasis Research Group, Imperial College School of Medicine, Hammersmith Hospital, Du Cane Road, London, W12 0NN, UK

Background. Thrombotic vascular occlusion resulting in infarction occurs during hyperacute rejection of allografts transplanted into sensitized patients and remains a major problem in experimental xenotransplantation. A similar process is also found in disorders of diverse etiology including atherosclerosis, vasculitis, and disseminated intravascular coagulation. Methods. We have previously constructed two membrane-tethered anticoagulant fusion proteins based on human tissue factor pathway inhibitor and the leech anticoagulant hirudin and demonstrated their functional efficacy in vitro. These constructs have now been modified by the addition of a P-selectin sequence to the cytoplasmic tail to localize them in Weibel-Palade bodies. They have been transfected into Weibel-Palade body-positive endothelial cells isolated from the inferior vena cava of normal pigs. Results. In resting endothelial cells, fusion protein expression colocalized with P-selectin and was confined to Weibel-Palade bodies. These cells had a procoagulant phenotype in recalcified human plasma. However, after activation with phorbol ester the anticoagulant proteins were rapidly relocated to the cell surface where they specifically inhibited the clotting of human plasma. Conclusions. Novel anticoagulant molecules may prove useful therapeutic agents for gene therapy in thrombotic disease and postangioplasty or for trans1

Department of Immunology. 2 MRC Clinical Sciences Centre Haemostasis Research Group. 3 Current address; Dept. of Medical Microbiology, Malmo¨ University Hospital, Lund University, S-205 02 Malmo¨, Sweden. 4 Address correspondence to: A. Dorling, Department of Immunology, Imperial College of Medicine, Hammersmith Hospital, Du Cane Road, London W12 0NN, UK.

genic expression in animals whose organs may be used for clinical xenotransplantation. Expression in vascular endothelial cells may be regulated by inclusion of P-selectin cytoplasmic sequence, to restrict cell surface expression to activated endothelium. Thrombotic vascular occlusion may occur in multisystem disorders, may complicate local procedures such as angioplasty and vascular surgery, and features prominently in the pathology of hyperacute rejection (HAR*) of allo- or xenografts, where it leads to catastrophic and irreversible infarction of the transplanted organ. In this setting, the procoagulant state within the graft develops from the simultaneous effects of complement activation (dependent on the deposition of anti-graft EC antibody), loss of proteoglycan-linked anticoagulant molecules such as antithrombin III and tissue factor pathway inhibitor (TFPI) from endothelial cell (EC) surfaces (1) and down-regulation of thrombomodulin, which is internalized after EC activation. In xenografts, another factor is the failure of some xenogeneic anticoagulant molecules to function efficiently across a species divide (2, 3) . Prevention of intravascular thrombosis by intraluminal expression of EC-tethered regulators of coagulation is an attractive strategy, particularly in the context of xenotransplantation, because widespread intravascular deposition of * Abbreviations: HAR, hyperacute rejection; ACTH, adrenocorticotrophin; DXR, delayed xenograft rejection; EC, endothelial cell; FITC, fluorescein isothiocyanate; FVIIa, activated clotting factor VII; FX/Fxa, clotting factor X/Xa; hTF/pTF, human or porcine tissue factor; IL-1a, interleukin-1a; IVC, inferior vena cava; mAb, monoclonal antibodies; PBS, phosphate-buffered saline; PMA, phorbol myristate acetate; TFPI, tissue factor pathway inhibitor; vWF, von Willebrand factor; WP, Weibel-Palade.