Liposome-Mediated Gene Transfection of Endothelial Nitric Oxide Synthase Reduces Endothelial Activation and Leukocyte Infiltration in Transplanted Hearts
Background—During cardiac ischemia-reperfusion injury, neutrophilic infiltration of the myocardium is mediated by adhesion molecule expression on activated coronary endothelium. Nitric oxide inhibits neutrophil adhesion to endothelium in vitro by blocking the nuclear factor (NF)-κB–dependent transcription of adhesion molecules. We investigated whether intraoperative gene delivery of endothelial nitric oxide synthase (eNOS) into donor hearts before transplantation would have a similar effect on an entire organ.
Methods and Results—In an allogeneic rabbit heart transplant model, liposomes complexed to the gene encoding eNOS were infused into the donor coronary circulation before transplantation. By 24 hours after transplantation, calcium-dependent nitrite production was significantly higher in eNOS-transfected donor hearts than in the 3 control groups: donor hearts transfected with empty plasmids alone, donor hearts treated with diluent only, and untransplanted native hearts. Intramyocardial neutrophil and T-lymphocyte populations were halved in eNOS-transfected hearts compared with control donor hearts (P<0.05). Moreover, the prevalence of NF-κB activation in microvascular endothelial cells and surrounding cardiac myocytes as well as endothelial vascular cell adhesion molecule-1 and intracellular adhesion molecule-1 expression were all significantly reduced in eNOS-transfected hearts compared with control donor hearts (P<0.01). Without immunosuppression, eNOS-transfected hearts survived longer than controls.
Conclusions—Intraoperative liposome-mediated gene delivery of eNOS to donor hearts can result in early gene expression sufficient to reduce ischemia-reperfusion injury by inhibiting NF-κB activation, adhesion molecule expression, and the early infiltration of leukocytes, all of which may improve graft survival.
Ischemia-reperfusion injury is the major cause of early mortality after cardiac transplantation and a risk factor in the >300 000 nontransplant cardiac surgical operations performed each year in the United States. Reperfusion after myocardial ischemia results in activation of coronary microvascular endothelial cell adhesion molecule expression, attraction of neutrophils to the myocardium, and subsequent myocardial injury.1 In clinical cardiac transplantation, the ischemic period of 1 to 4 hours between donor heart procurement and implantation affords a unique opportunity to intervene in this chain of events before reperfusion. Similarly, during routine cardiac surgery, the aortic cross-clamp time during cardiopulmonary bypass offers a similar period during which an intraoperative intervention could be performed before reperfusion of the heart.
Nitric oxide (NO) has been shown, both in vitro and in vivo, to inhibit cytokine-induced endothelial expression of adhesion molecules (vascular cell adhesion molecule-1 [VCAM-1], intracellular adhesion molecule-1 [ICAM-1], and E-selectin) and proinflammatory cytokines by the induction, stabilization, and nuclear translocation of IκBα, which, in turn, blocks activation of the proinflammatory transcription factor nuclear factor-κB (NF-κB).2 3 Because leukocyte recruitment to activated endothelium is dependent on adhesion molecule receptor-ligand interactions, gene therapy approaches that increase NO production might inhibit donor endothelial adhesion molecule expression and reduce host leukocyte infiltration into the transplanted organ. The strategic questions to address here are whether our currently inefficient gene delivery systems can produce sufficient gene product to have a therapeutic effect on an entire organ and whether gene expression can be achieved within the short time frame necessary to mitigate very early injury patterns.
Because NO can diffuse to surrounding cells, gene delivery of NO synthase (NOS) might be effective even if relatively few cells expressed the gene product. We used the gene encoding endothelial (eNOS) rather than inducible (iNOS) NOS because its calcium dependency and potential for feedback control of iNOS production3 might allow some homeostatic physiological control as protection against NO overproduction. Our laboratory has previously shown that intraoperative ex vivo delivery of cationic liposomes to the donor heart before transplantation is quite feasible4 and results in histological evidence of reporter gene expression in the donor heart by at least 6 hours after transfection. Because we and others have also shown that de novo production of the integrin adhesion molecules after transplantation requires a similar window of time,5 6 gene products may be produced within enough time to mitigate even ischemia-reperfusion injury.
We recently reported that liposome delivery of reporter genes to rabbit carotid arteries targets primarily endothelial cells; gene product expression, evident histologically in 3% of endothelial cells, was still persistent at 21 days.7 When liposome gene delivery of eNOS was applied to transplanted carotids, the prevalence of eNOS-expressing cells and NO production was doubled, endothelial NF-κB activation and adhesion molecule expression was suppressed, and vessel wall iNOS expression decreased 10-fold. ENOS gene delivery significantly reduced neointima formation in these transplanted arteries.7
To assess the efficacy of intraoperative gene delivery in cardiac transplants, in this study we assayed eNOS-transfected donor hearts 24 hours after transplantation for evidence of NO production, NF-κB activation, endothelial ICAM-1 and VCAM-1 expression, and neutrophil and T-lymphocyte infiltration. Finally, we asked whether single-dose pretransplant gene transfection would extend cardiac graft survival without other immunosuppression.
New Zealand White rabbits and Stauffland rabbits (R and R Rabbitry, Stanwood, Wash) were used for all experiments. All experiments were performed in accordance with the Principles of Laboratory Animal Care, National Society for Medical Research, and the Guide for the Care and Use of Laboratory Animals, Institute of Laboratory Animal Resources, National Academy of Science (NIH publication 86-23, revised 1985).
Our model of heterotopic cervical heart transplantation between 2 outbred strains of rabbits produces well-characterized early neutrophil infiltration as well as allogeneic transplant cellular and vascular rejection.5 Briefly, both donor and recipient animals were anesthetized with a mixture of ketamine (35 mg/kg), xylazine (5.0 mg/kg), and atropine (0.032 mg/kg); intubated; and ventilated with halothane in 100% O2. After systemic heparinization (300 U/kg), donor hearts were arrested with cold (4°C) nonionic Stanford cardioplegia solution. Three milliliters of liposome-DNA complexes diluted in 10 mL of D5W (final concentration of 0.45 mg DNA/mL) was injected into the aortic roots of the donor hearts to perfuse the coronary circulation. At implantation, the donor ascending aorta and pulmonary artery were anastomosed to the recipient carotid artery and jugular vein, respectively. Before reperfusion, the coronary circulation was flushed with heparinized D5W to remove excess, nonadherent liposomes. The ischemic time of the donor hearts from procurement to reperfusion after implantation averaged 32±5 minutes (mean±SEM). Recipient rabbits were allowed to fully recover and then were euthanized 24 hours or 7 days later for assessment of both the transplanted (donor) and the native (recipient) hearts.
In the graft survival experiments, transplanted hearts were assessed daily for heartbeat until asystole. No immunosuppression was given.
Preparation of Liposome-DNA Complexes
The transfected plasmids were either a pUC19 control plasmid without a functional gene or the plasmid carrying a chloramphenicol acetyltransferase (CAT) reporter gene, or a human eNOS–expressing DNA clone. Liposomes (Valentis Inc) and cationic lipid/DNA complexes were prepared as described previously,8 9 producing a final concentration of 1.5 mg/mL plasmid DNA, 1.5 mmol/L cholesterol, and 1.5 mmol/L 1-[2-[9-(Z)-octadecenoyloxy]]-2-(Z)-heptadecenyl]-3-[hydroxyethyl] imidazolinium chloride (BODAI).
With regard to reporter gene pilots, we have previously shown that liposome-CAT gene delivery to donor hearts results in CAT gene expression in all myocardial distributions at 24 hours after transplantation.4 Here, 7 additional donor hearts treated with liposome-CAT complexes were examined 7 days after transplantation; CAT enzyme activities in homogenized specimens from donor and native hearts were assayed by liquid scintillation scanning and reported as counts per minute per milligram total protein (cpm/mg).4
Eighteen donor hearts were perfused with liposomes complexed to the eNOS gene: 10 were excised at 24 hours after transplantation, 2 at 7 days for histology, and 6 followed to graft asystole. Three different sets of controls were performed for comparison with 24-hour eNOS-treated hearts. Five donor hearts were treated with the same liposomes complexed to the pUC19 plasmid and promoter sequence without a functional gene; 5 donor hearts were treated with the diluent, D5W, alone; and 5 untransplanted native hearts, without ischemia or reperfusion injury, served as an internal control for the effects of systemically circulating cytokines. Eight other transplanted control hearts, without liposome gene delivery, were followed out to graft asystole (n=6) or excised at 7 days (n=2) for histology.
To quantify gene product expression by vascular cells alone, without surrounding myocardium, nitrite production was measured by the Griess reaction7 in 5×5-mm specimens from the aortic root at the coronary ostia and the coronary sinus of each heart as examples of coronary inflow and outflow. Specimens were incubated in the presence or absence of calcium ionophore (ionomycin, Sigma Chemical Co, 10−5 mol/L for 2 hours at 37°C) to differentiate calcium-dependent eNOS gene expression from calcium-independent iNOS expression.
Quantification of End Points on Histology
Hearts were prepared for histological examination by cutting serial 5-μm-thick transverse sections at the mid–papillary muscle level and fixation in 4% formalin (zinc-buffered fixative7 was used for ICAM-1 and VCAM-1 detection). After fixation, tissues were embedded in paraffin. As previously described,5 myocardial cross sections were oriented on a defined grid. An image analysis system (Optimus Corp) was used to calibrate type-specific cell populations in 18 standard fields of myocardium for each cross section, viewed at 200-fold magnification (400-fold magnification for NF-κB quantification). Data presented for each heart represent the mean values of the 18 fields per cross section averaged over the 5 to 6 such sections evaluated per heart.
The prevalences of NF-κB activation and ICAM-1 and VCAM-1 expression in endothelial cells were assessed quantitatively within myocardial specimens by counting the vessels that had any positive endothelial staining divided by the total number of vascular profiles (defined by platelet and endothelial cell adhesion molecule-1, CD31, staining). Vessels in which immunostaining was present but not circumferential were counted as positive for this analysis. NF-κB activation of cardiac myocytes was calculated as the percentage of myocytes positively stained for activated p65 divided by the total number of myocyte nuclei. NF-κB activation and ICAM-1 expression on infiltrating host leukocytes were not evaluated because treatment was directed only to the donor organ endothelium. T-lymphocyte infiltration was quantified by computerized image analysis of immunocytochemistry specimens; neutrophils were counted manually on the same grid system; leukocyte data are presented as the mean number of cells per field.
Immunocytochemistry was used to identify infiltrating T lymphocytes, endothelial VCAM-1 and ICAM-1 expression, and NF-κB activation. As previously,7 the antibodies used were directed against rabbit T lymphocytes, monoclonal antibody (mAb) Tib 188 (1:500 dilution, American Type Culture Collection), rabbit VCAM-1, mAb Rb 1/9 (1:100 dilution), and rabbit ICAM-1, mAb Rb 2/3 (1:500 dilution) (both gifts from Dr Myron Cybulsky). Activated NF-κB was detected with a monoclonal antibody to an epitope on the p65 subunit (1:50 dilution; anti–NF-κB–p65, Roche Molecular Biochemicals) that is exposed only on nuclear translocation and thus binds only to the active, nuclear form of NF-κB .
Briefly, after deparaffinization and dehydration, all sections were incubated in methanol/3% peroxidase 1:1 for 10 minutes to block endogenous peroxidase, then in blocking solution (1% BSA/PBS, 1% horse serum, and 0.1% Triton X-100) for 30 minutes, followed by incubation with primary antibodies for 1 hour at room temperature. After washing with PBS, all sections were incubated with biotinylated horse anti-mouse IgG (Vector Laboratories) for 30 minutes, washed, then treated with avidin-peroxidase (ABC Elite kit, Vector Laboratories) for 30 minutes. The reaction was developed with DAB substrate (Vector Laboratories). Finally, sections were counterstained with hematoxylin to quantify the total number of nuclei per field as the denominator for the calculations above. Antigen unmasking was required for NF-κB detection, accomplished after alcohol dehydration by placing sections in a 0.1 mol/L citrate buffer bath (pH 6.0), which was then heated for 10 minutes in a microwave oven.
Neutrophils were identified by esterase staining (naphthol AS-D chloroacetate esterase procedure, Sigma).
Paraffin sections from 2 eNOS-transfected and 1 control D5W-treated donor hearts were examined for apoptosis by terminal deoxynucleotidyl transferase–mediated dUTP nick end-labeling (TUNEL) (In Situ Cell Death Kit, Roche Molecular Biochemicals) to label DNA strand breaks. Apoptotic cells in vascular profiles and cardiac myocytes were evaluated separately.
Continuous variables are presented as the mean±SEM; comparisons between groups were performed with Student’s t test; differences were considered to be statistically significant at a value of P<0.05.
Reporter Gene Pilots
In 7-day pilots, mean CAT enzyme activities in homogenized right and left ventricular myocardial specimens from the liposome-CAT–treated, transplanted donor hearts were 5- to 7-fold higher than the baseline “negative” readings in native hearts. Consistent with vascular endothelial cells as the primary transfection targets, reporter gene expression was reliably persistent in all aortic (coronary ostial) specimens at 7 days but 10-fold lower than in similar specimens examined at 24 hours.4 At 7 days, mean CAT activities were 36±4.8 versus 4.7±1.1×103 cpm/mg in the treated and native aortic specimens, respectively (P<0.001).
We observed significantly higher production of nitrite in aortic (coronary ostia) and coronary sinus specimens obtained from eNOS-transfected hearts than from control hearts treated with liposome-plasmid complexes without a functional gene or from untransplanted native hearts (P<0.01, Figure 1⇓). Given the rapid first-pass adherence of cationic liposomes to endothelium,4 9 we observed higher levels of nitrite production near the delivery site in the aorta than in the coronary sinus (414±66 versus 149±25 nmol · L nitrite−1 · mg tissue−1, P<0.01) in the eNOS-transfected hearts. Furthermore, in eNOS-transfected hearts, nitrite production increased 4-fold after calcium ionophore treatment. This calcium dependence is strong evidence that the majority of NO was produced by eNOS expressed from the transfected gene. In these vascular specimens, there were no other significant differences in endogenous calcium-dependent or -independent nitrite production between liposome-transfected control donor hearts and native hearts.
The prevalence of NF-κB activation (NF-κB nuclear localization) in intramyocardial coronary endothelial cells was significantly reduced in eNOS-transfected myocardial specimens compared to pUC-transfected control specimens and to diluent (D5W)-treated controls (27±3% versus 87±3% and 70±4%, respectively, P<0.01) (Figures 2⇓ and 3a⇓). NF-κB activation in eNOS-treated, transplanted hearts was still higher than in untransplanted native hearts (P<0.05). Among cardiac myocytes, the prevalence of NF-κB activation was 65% less in the eNOS-transfected hearts than in control donor hearts (19±2% versus 55±3% and 55±4%, respectively, P<0.01) (Figure 3b⇓).
Endothelial ICAM-1 and VCAM-1 Expression
The percentage of intramyocardial vascular profiles with endothelial expression of ICAM-1 and/or VCAM-1 expression were both significantly reduced in eNOS-treated hearts compared with control hearts treated with liposome-pUC plasmid complexes and those treated with the diluent, D5W, alone (ICAM-1: 46±5% versus 92±1% and 80±3%, respectively; VCAM-1: 47±14% versus 92±1% and 91±6%) (Figures 4e⇓ through 4h⇓ and 5⇓). Adhesion molecule expression in transplanted eNOS-transfected hearts, however, was still higher than in untransplanted native hearts (endothelial ICAM-1, 2±0.5%; endothelial VCAM-1, 2±1%, respectively).
Leukocyte Infiltration at 24 Hours
The number of neutrophils per field was 50% less in eNOS-transfected myocardium (9±1 cells/field) compared with pUC-transfected myocardium (18±1 cells/field, P<0.01) and control myocardium treated with diluent only (16±4.3 cells/field, P<0.05) (Figures 4a⇑ and 4b⇑ and 6a⇓.
A significant reduction in T-lymphocyte infiltration was also demonstrated in eNOS-transfected hearts compared with hearts transfected with the pUC control plasmid (89±16 versus 172±21 cells/field, P<0.01) as well as hearts treated with the diluent, D5W, alone (155±12 cells/field, P<0.01) (Figures 4c⇑ and 4d⇑ and 6b).
The prevalence of apoptosis was examined to ascertain whether increased NO production would cause apoptosis. In both the eNOS-transfected and control donor hearts, <1% of vascular endothelial and smooth muscle cells had evidence of apoptosis by TUNEL at 24 hours after transplantation. Among cardiac myocytes, only 4% of myocytes in the eNOS-transfected hearts had positive TUNEL staining, compared with 29% in the untreated control donor hearts.
Histology at 7 Days
At 7 days after transplantation, 2 eNOS-transfected transplanted hearts evidenced markedly less vascular rejection than untreated control hearts (19% versus 70% of arterial profiles with vascular rejection, respectively; Figure 7⇓). eNOS-transfected hearts also had fewer T lymphocytes (mean 161 versus 288 cells/field) and macrophages (mean 15 versus 27 cells/field) present at this time point. Only rare neutrophils were seen in either group.
Without immunosuppression, donor hearts transfected with a single dose of liposome-eNOS complexes before transplantation survived significantly longer (13.5±1.3 days) than control donor hearts without liposome-gene transfection (9.3±1.8 days, P<0.05) (Figure 8⇓).
Adhesion of host leukocytes to donor endothelium is the first step in ischemia-reperfusion injury as well as in graft cellular rejection. We have previously shown that antibody blockade of adhesion molecule receptors on recipient leukocytes can markedly reduce both neutrophil and T-lymphocyte infiltration of the transplanted donor heart during ischemia-reperfusion injury5 and early graft rejection10 in this model. Host leukocyte alteration, however, like other existing immunosuppression, impairs host defenses against infection. In these studies, we investigate whether the same end point of adhesion molecule blockade could be achieved by modification of the donor coronary endothelial adhesion ligands instead of the recipient leukocyte receptors.
These experiments demonstrate that liposome gene delivery of eNOS, even given current transfection inefficiencies, was sufficient to reduce activation not only of the transfected endothelial cells but also of the surrounding myocytes. Also, by using perfusion delivery techniques and endothelial cell transfection, which may be more physiological, we did not see the myocardial apoptosis recently reported with a gene therapy strategy that employed direct myocardial injection of other eNOS-liposome vectors.11
Previous studies have demonstrated that myocardial ischemia-reperfusion injury is exacerbated in the absence of eNOS12 and that inhaled NO13 or infusion of NO donors will ameliorate ischemia-reperfusion injury in several models.14 15 16 Organ-specific gene therapy approaches may make it possible to generate therapeutic local levels of NO at target tissue sites without incurring systemic side effects. Several previous in vivo studies, using both liposome17 and adenoviral18 19 vectors, have used arterial gene transfer of eNOS to inhibit neointimal formation in nontransplant settings. Known beneficial effects of eNOS on coronary vasodilatation, platelet function, endothelial cell survival20 and physiology,21 vascular wall activation,22 and cell-regulatory events23 may be operative in the amelioration of early coronary endothelial activation and later vascular rejection in this transplant model.
The fact that vascular nitrite production was increased only in eNOS-transfected specimens, with little contribution of endogenous nitrite production, supports the premise that gene transfer was responsible for the observed results. Any concomitant detrimental24 (or even cardioprotective) effects of myocyte and/or macrophage iNOS were controlled for in the liposome-transfected, transplanted hearts without functional genes.
Although not statistically significant, it was of interest that higher mean prevalences of NF-κB activation, adhesion molecule expression, and leukocyte infiltration were seen in the control hearts transfected with the liposome-pUC plasmid complexes than in the hearts treated with D5W alone. This might represent an activation phenomenon induced either by the liposomes25 or, more likely, by the bacterial plasmid DNA as a result of unmethylated CpG motifs,26 which can contribute to immune activation, immediate NF-κB activation, and inflammatory gene induction.27 If so, the NO produced by the transduced eNOS presumably also downregulated this mild activation induced by the control liposome-plasmid complexes, an important safety point for clinical applications.
In summary, these studies demonstrate that intravascular liposome delivery of the eNOS gene to donor hearts leads to measurable endothelial expression of NO after transplantation, reductions in early endothelial activation and leukocyte infiltration, and extended graft survival without immunosuppression. This simple intraoperative technique may herald the clinical application of gene therapy as adjunctive therapy in the transplantation of hearts and other solid organs as well as in routine cardiac surgery outside the realm of transplantation.
This work was supported by the American Heart Association, grant 9650559N. The authors would like to acknowledge Dr Jeanette Ennis for reviewing the manuscript and the Valentis Corporation for providing the liposome complexes used in this study.
- Received November 7, 2000.
- Revision received January 28, 2001.
- Accepted January 29, 2001.
- Copyright © 2001 by American Heart Association
Jordan JE, Zhao ZQ, Vinten-Johansen J. The role of neutrophils in myocardial ischemia-reperfusion injury. Cardiovasc Res. 1999;43:860–878.
Peng HB, Libby P, Liao JK. Induction and stabilization of I kappa B alpha by nitric oxide mediates inhibition of NF-kappa B. J Biol Chem. 1995;270:14214–14219.
Katsuyama K, Shichiri M, Marumo F, et al. NO inhibits cytokine-induced iNOS expression and NF-kappa B activation by interfering with phosphorylation and degradation of I kappa B-alpha. Arterioscler Thromb Vasc Biol. 1998;18:1796–1802.
Iwata A, Sai S, Moore M, et al. Gene therapy of transplant arteriopathy by liposome-mediated transfection of endothelial nitric oxide synthase. J Heart Lung Transplant. 2000;10:1017–1028.
McLean JW, Fox EA, Baluk P, et al. Organ-specific endothelial cell uptake of cationic liposome-DNA complexes in mice. Am J Physiol. 1997;273:H387–H404.
Kawaguchi H, Shin WS, Wang Y, et al. In vivo gene transfection of human endothelial cell nitric oxide synthase in cardiomyocytes causes apoptosis-like cell death: identification using Sendai virus–coated liposomes. Circulation. 1997;95:2441–2447.
Jones SP, Girod WG, Palazzo AJ, et al. Myocardial ischemia-reperfusion injury is exacerbated in absence of endothelial cell nitric oxide synthase. Am J Physiol. 1999;276:H1567–H1573.
Lefer DJ, Nakanishi K, Vinten-Johansen J. Endothelial and myocardial cell protection by a cysteine-containing nitric oxide donor after myocardial ischemia and reperfusion. J Cardiovasc Pharmacol. 1993;22:S34–S43.
Pabla R, Buda AJ, Flynn DM, et al. Nitric oxide attenuates neutrophil-mediated myocardial contractile dysfunction after ischemia and reperfusion. Circ Res. 1996;78:65–72.
von der Leyen HE, Gibbons GH, Morishita R, et al. Gene therapy inhibiting neointimal vascular lesion: in vivo transfer of endothelial cell nitric oxide synthase gene. Proc Natl Acad Sci U S A. 1995;92:1137–1141.
Varenne O, Pislaru S, Gillijns H, et al. Local adenovirus-mediated transfer of human endothelial nitric oxide synthase reduces luminal narrowing after coronary angioplasty in pigs. Circulation. 1998;98:919–926.
Janssens S, Flaherty D, Nong Z, et al. Human endothelial nitric oxide synthase gene transfer inhibits vascular smooth muscle cell proliferation and neointima formation after balloon injury in rats. Circulation. 1998;97:1274–1281.
Murohara T, Witzenbichler B, Spyridopoulos I, et al. Role of endothelial nitric oxide synthase in endothelial cell migration. Arterioscler Thromb Vasc Biol. 1999;19:1156–1161.
Ooboshi H, Toyoda K, Faraci FM, et al. Improvement of relaxation in an atherosclerotic artery by gene transfer of endothelial nitric oxide synthase. Arterioscler Thromb Vasc Biol. 1998;18:1752–1758.
Qian H, Neplioueva V, Shetty GA, et al. Nitric oxide synthase gene therapy rapidly reduces adhesion molecule expression and inflammatory cell infiltration in carotid arteries of cholesterol-fed rabbits. Circulation. 1999;99:2979–2982.
Sharma RV, Tan E, Fang S, et al. NOS gene transfer inhibits expression of cell cycle regulatory molecules in vascular smooth muscle cells. Am J Physiol. 1999;276:H1450–H1459.
Arstall MA, Sawyer DB, Fukazawa R, et al. Cytokine-mediated apoptosis in cardiac myocytes: the role of inducible nitric oxide synthase induction and peroxynitrite generation. Circ Res. 1999;85:829–840.
Freimark BD, Blezinger HP, Florack VJ, et al. Cationic lipids enhance cytokine and cell influx levels in the lung following administration of plasmid:cationic lipid complexes. J Immunol. 1998;160:4580–4586.