The role of inducible nitric oxide synthase in cardiac allograft rejection

Time for primary review 29 days. Cardiac transplantation is an effective therapy for end-stage heart failure with one and five year survival rates about 80% and 65% respectively [1]. Despite these good results, cardiac allograft rejection remains a problem which produces impaired ventricular performance and death of cardiac myocytes ultimately causing congestive heart failure, low cardiac output, hypotension and reduced pressor responses to catecholamines. Chronic allograft rejection also results in the development of transplant-associated coronary atherosclerosis, a vasculopathy that produces myocardial ischemia, infarction and sudden death and is the leading cause of death in cardiac transplant recipients after the first year [2]. Although there is a large body of evidence concerning the immunological interactions involved in cardiac allograft rejection, knowledge concerning the cellular and biochemical mechanisms responsible for contractile dysfunction and for death of heart muscle cells remains incomplete. Since an inflammatory reaction in the myocardium is an intrinsic component of the pathological changes observed during cardiac allograft rejection, our laboratory undertook an investigation of the participation of nitric oxide synthases, particularly the inducible nitric oxide synthase (iNOS), in the manifestations of cardiac allograft rejection. The five electron oxidation of l-arginine to l-citrulline and biologically active nitric oxide (NO) is important to a large variety of physiological and pathological processes [3, 4]. NO synthesis is accomplished by the three isoforms of NOS, the neuronal NOS (nNOS, NOS 1) originally identified in brain, inducible NOS (iNOS, NOS 2) originally identified in macrophages and endothelial NOS (eNOS, NOS 3) originally identified in endothelial cells. Constitutive nitric oxide synthases (nNOS and eNOS) require calcium and calmodulin as cofactors and generate low amounts of NO. Small amounts of NO released by endothelial cells in response to hormones or sheer stress interact with soluble guanylyl cyclases to increase the formation of cyclic GMP in target cells such as platelets, endothelial cells and vascular smooth muscle cells promoting inhibition of platelet adhesion and aggregation, inhibition of leukocyte adhesion and migration and vasodilation respectively. The NO produced by nNOS acts as a neurotransmitter in brain, in cells of the nonadrenergic, noncholinergic nervous system and in skeletal muscle [5]. The inducible NO synthases expressed in macrophages, endothelial cells, vascular smooth muscle cells and cardiac myocytes in response to cytokines (such as IL-1β, TNF-α, IFN-γ, IL-6) or bacterial endotoxin do not require calcium and calmodulin as cofactors and generate substantially larger amounts of NO for long periods of time [3–6]. NO produced by iNOS in activated macrophages is cytotoxic and participates in their antimicrobial actions [4, 7]. NO produced by vascular smooth muscle cells and cardiac myocytes in response to endotoxin has been implicated in the pathogenesis of hypotension in association with bacterial infections [4]. The continuous large production of NO by iNOS in cardiac myocytes has been shown to influence myocardial contractile responses and to affect heart rate [6]. Other factors important in the synthesis of NO by iNOS include the availability of the substrate, l-arginine, and of the cofactors NADPH and tetrahydrobiopterin (THB4) [3–7]. When iNOS is induced by cytokines in macrophages, endothelial cells or cardiac myocytes, cationic amino acid transporter proteins, CAT 1, CAT 2B (both high affinity) and CAT 2A (low affinity) are coinduced which can increase the intracellular l-arginine concentration. GTP cyclohydrolase, the key enzyme in THB4 biosynthesis, is also coinduced with iNOS. Other experimental evidence obtained with nNOS and iNOS indicates that NO can exert a negative feedback on NO synthesis [6]. In activated macrophages NO can inhibit iNOS activity by reducing the availability of heme and its insertion into monomers, blocking formation of the dimers required for enzyme activity. During cardiac allograft rejection there is significant release of cytokines by activated T lymphocytes and macrophages which participate in the immune response to foreign HLA and other antigens present in endothelial and other cells of the transplanted heart [8]. Because cytokines, particularly those released by the Th1 subset of lymphocytes, were reported to stimulate iNOS expression in several cell systems, Yang et al. used a rat model to investigate iNOS expression during cardiac allograft rejection [9]. Heterotopic abdominal cardiac transplantation was performed using Lewis strain donor hearts and Wistar–Furth strain recipients. Lewis to Lewis transplants were performed as syngeneic controls. Allograft rejection is complete at days 6–7 with this model. Cardiac allografts examined at day 5 manifested reduced contractility and histological changes of severe rejection, with an inflammatory myocardial infiltrate composed of lymphocytes and macrophages, edema and damage and destruction of cardiac muscle fibers. The mRNA for iNOS and iNOS protein were detected in ventricular homogenates and also in isolated purified myocytes from the rejecting cardiac allografts but not in ventricular tissue or myocytes from the syngeneic control grafts. iNOS enzyme activity and tissue cyclic GMP levels were also increased significantly in the rejecting allografts. Immunostaining with a highly specific antibody indicated that iNOS protein was present in macrophages infiltrating the myocardium and also in cardiac myocytes of the rejecting allografts. Some lymphocytes and microvascular endothelial cells in the allografts also showed positive iNOS immunostaining. The data indicated that myocardial iNOS mRNA, protein and enzyme activity are induced in infiltrating macrophages and cardiac myocytes of rejecting allografts. Using a similar Lewis to ACI strain heterotopic rat cardiac transplant model Worrall and coworkers confirmed that iNOS mRNA and enzyme activity were increased in rejecting cardiac allografts [10]. In agreement with a previous report serum nitrite/nitrate levels were also significantly increased in the allograft recipients [10–12]. Immunostaining revealed iNOS protein in infiltrating macrophages but not in cardiac myocytes. Whether absent iNOS immunostaining in cardiac myocytes in the Worrall studies [10, 11]relates to rat strain differences, milder rejection or methodological problems is unclear. Treatment with aminoguanidine, a drug which inhibits iNOS activity (but which also has antioxidant properties, reduces advanced glycation end product formation and inhibits diamine oxidase and aldose reductase) improved allograft papillary muscle contractile performance, reduced the extent of pathological changes and prolonged allograft survival from 10 to 15 days [10, 11]. EPR signals in myocardial tissue from the rejecting grafts were indicative of the formation of nitrosylferromyoglobin and/or nitrosylferrohemoglobin and of nonheme iron–dinitrosyl complexes; these signals were also ameliorated by aminoguanidine [10]. Russell et al., using a Lewis-F344 rat cardiac transplantation model, also observed a prolonged increase in expression of iNOS mRNA and immunostaining of myocardial macrophages and of vascular smooth muscle cells in rejecting cardiac allografts [13]. In studies of serial endomyocardial biopsies of patients during the first year following heart transplantation, Lewis and coworkers found iNOS mRNA during the first 180 days [15]. iNOS protein was demonstrated by immunohistochemistry in cardiac myocytes. There was also increased myocardial cGMP. iNOS mRNA expression was not related to the ISHLT histological grade of rejection but was associated significantly with systolic and diastolic dysfunction of the left ventricle. Winlaw et al. in a study of Brown–Norway to Lewis allografts and Lewis–Lewis isografts found that urinary nitrite excretion in untreated allograft rejection was increased 8 fold above basal excretion of isografts and that peak nitrate excretion occurred early in rejection and was attenuated by immunosuppressive therapy [16]. Allograft associated vasculopathy is considered to be a form of chronic rejection [2]. Transplant associated coronary artery disease tends to be diffuse, to extend to small arteries and arterioles and to involve a proliferative intimal response of smooth muscle cells and to a lesser extent of macrophages. Russell et al. and Akyurek et al. reported that eNOS was present in endothelial cells and that iNOS expression was upregulated in macrophages and smooth muscle cells in the intimal lesions found in rat models [14, 17]. Smooth muscle immunostaining was particularly apparent at later time points in lesions from the rat. Ravalli et al. used in-situ hybridization and immunostaining to examine iNOS expression in human transplant coronary artery disease [18]. Both iNOS mRNA and protein were detected in neointimal macrophages, neointimal ‘foam’ cells and neointimal smooth muscle cells. Immunostaining for nitrotyrosine was also found in the same cell types. These findings are similar to those found in human atherosclerotic lesions and in mouse models of the disease [19, 20]. Whether iNOS expression is pathogenetic in promoting the vascular disease (e.g. by increasing oxidant stress and the expression of oxidant sensitive genes) or is protective against the development of vascular disease is unclear. In the study of Aji et al. the administration of l-arginine supplements to LDL receptor knockout mice fed a high cholesterol diet reduced xanthoma formation and the extent of atherosclerotic aortic lesions by 40% [20]. This beneficial effect was abrogated when a NOS inhibitor was coadministered with the l-arginine supplement suggesting that NO produced by eNOS and iNOS was responsible for lesion reduction. NO is known to inhibit platelet and white cell adhesion and transmigration of white blood cells across the endothelium and to block smooth muscle cell proliferation [3, 4, 6]. Another possibility is that NO contributes to vascular remodeling by its known effects to promote apoptosis of macrophages and vascular smooth muscle cells [21, 22]. A recent preliminary report which indicated that transplant associated vasculopathy was more extensive in allografts in iNOS deficient mice is also consistent with a protective effect of iNOS in this disorder [23]. Cellular immune responses constitute major components of acute and chronic allograft rejection [24]. Activated T lymphocytes, both CD4+ and CD8+, along with macrophages and endothelial cells expressing class II MHC antigens and multiple cytokines released by macrophages and other cells (including IL-1, IL-2, IFN-γ, TNF-α, IL-6 and IL-10) are present in the myocardium and vessels undergoing allograft rejection [8, 25]. iNOS expression occurs early in the rejection process as reflected by immunostaining of myocardial tissues and also by urinary excretion of nitrite/nitrate [9–11, 16]. The administration of immunosuppressive therapy such as dexamethasone, cyclosporine A and FK506 alone or in combination resulted in an amelioration of the pathological changes in the myocardium and also a reduction of the levels of iNOS mRNA, and (in another report) a delay in the onset and peak rise in urinary excretion of nitrite/nitrate [9–11, 16, 26]. As mentioned previously, the administration of aminoguanidine, a somewhat selective iNOS inhibitor, also resulted in an improved survival and a delay in the histological appearance of severe rejection [10, 11]. These data suggest that iNOS in macrophages and cardiac myocytes plays a role in the pathogenesis of rejection but this is not exclusive or necessarily the most important role. The specific signal which induces iNOS expression during allograft rejection has not been defined. Nevertheless, it is probable that cytokines produced by activated macrophages play an important role [8, 25]. Interferon-γ, TNF-α and IL-1β have been reported to induce iNOS mRNA, protein and enzyme activity in macrophages, vascular smooth muscle cells and isolated rat cardiac myocytes in vitro [3, 4, 6]. It is also probable that interaction between the CD40 ligand expressed on CD4+ T-lymphocytes and its target molecule CD40 also plays a role. Interaction by cell contact between the CD40 ligand on T-cells and CD40 on macrophages and has been reported to induce iNOS in vitro [27]. In recent studies of human coronary arteries using immunohistochemistry, Szabolcs et al. found that in both transplant coronary artery disease and atherosclerosis CD40 ligand+ T-lymphocytes were present in the intimal lesions along with abundant positive immunostaining for CD40 on endothelial cells, macrophages, ‘foam’ cells and smooth muscle cells [28]. Larson et al. using a mouse model of heterotopic cardiac transplantation reported that treatment with an antibody to CD40 ligand (anti-gp39) at the time of transplantation markedly prolonged graft survival [29]. Allografts from treated recipients showed decreased expression of mRNA for iNOS but unaltered expression of transcripts for T-cell cytokines or the costimulatory molecules B7-1, B7-2 [29]. In a subsequent publication this group demonstrated that simultaneous therapy which blocked CD40 ligand-CD40 interactions with an antibody which inhibited CD28-B7 interactions aborted T-cell clonal expansion in vitro and in vivo and resulted in long term acceptance of cardiac and skin allografts and inhibition of the development of coronary vascular disease in the transplanted hearts [30]. Impaired ventricular contractile performance and cardiac myocyte death are the hallmarks of cardiac allograft rejection. Much evidence suggests that NO produced by iNOS contributes to allograft contractile dysfunction. NO acts upon soluble guanylyl cyclase to form cyclic GMP which is known to decrease cardiac myocyte L-type calcium channel current and the contractile responses of cardiac myofilaments to calcium [6, 31]. Increased myocardial cGMP has been found in guinea pigs treated with endotoxin and in rat and human cardiac allografts [9, 15, 32]. In studies by Brady et al. the amplitude of contraction of isolated guinea pig cardiac myocytes was reduced in cells from animals treated with endotoxin to induce iNOS and in myocytes from normal guinea pigs that were treated with media containing NO or the NO donor drug sodium nitroprusside [32, 33]. In studies of isolated rat cardiac myocytes by Balligand, the induction of iNOS in the myocytes by cytokines was not associated with a depression of basal contractile function but with a reduction in the contractile responses to β adrenergic agonists that could be reversed by administration of NOS inhibitors [34, 35]. Worrall et al. reported that iNOS derived NO in early rejection of cardiac allografts (prior to widespread cell death) was associated with impaired contractility of isolated papillary muscles at baseline and during β adrenergic, adenylate cyclase and calcium stimulation [10, 11, 36]. There were also associated membrane dysfunctions [36]. The impaired contractile responses were ameliorated by aminoguanidine or therapy with corticosteroids which inhibited iNOS expression. As mentioned previously, in studies of patients following cardiac transplantation Lewis et al. found that iNOS mRNA was inversely related to cardiac function, i.e. ejection fraction [15]. Paulus et al. found that in human allografts, iNOS mRNA (determined by RT-PCR in endomyocardial biopsies) was an independent variable relating a reduced peak contractile response to dobutamine to simultaneous shortening of systole, i.e. there was earlier onset of left ventricular relaxation and reduction of left ventricular end-systolic pressure [37]. The mechanisms by which NO produced by iNOS influences ventricular performance remain largely unexplored but are potentially multiple [6]. In addition to increasing cGMP, NO can produce auto-ADP ribosylation of glycolytic enzymes, inhibition of ribonucleotide reductase, and activation of poly ADP ribose synthetase which can lead to depletion of cellular energy stores. NO can also inhibit enzymes involved in the mitochondrial reducing and can inhibit the acid enzyme NO and its such as can with to form which can in with other containing their activity. the product of the interaction of NO and is a oxidant which can cause of and and to form can also produce of in cell of contractile such as the contractile function of cardiac myofilaments As be NO in large also to death of cardiac myocytes. of cardiac myocytes is the of cardiac allograft rejection. Because iNOS expression was apparent in macrophages and cardiac myocytes in rejecting allografts and the of NO by iNOS in macrophages is important in the immune against and other et al. performed in vitro to investigate effects of NO in the heart In the first macrophages were with cytokines to induce iNOS. and control macrophages were with isolated rat cardiac myocytes using to contact between the cell types. There was significantly myocyte death by release and by in with macrophages expressing iNOS. The death rate to control levels when with activated macrophages were treated with which inhibits NOS enzyme activity. The data indicated that NO produced by iNOS in activated macrophages can be to cardiac myocytes. In showed that rat cardiac myocytes treated with TNF-α, IL-1β and expressed iNOS mRNA, protein and enzyme activity and also a death rate control myocytes The increased death of cardiac myocytes was inhibited by the administration of to block NO formation and by of the myocytes with which reduced iNOS expression. The data that cytokines can induce iNOS expression in cardiac myocytes and that NO produced can be to the cardiac myocytes. the that NO produced by macrophages infiltrating the myocardium or by cardiac myocytes in a is potentially cytotoxic to heart muscle cells. be not in allograft rejection but in other pathological such as myocardial infarction or in which macrophages and cytokines are abundant and in which there is myocardial expression of iNOS. Although there is concerning immune mechanisms of cardiac myocyte knowledge concerning mechanisms of myocyte death during cardiac allograft rejection is [24]. During advanced of rejection there can be of heart muscle cells. This is by of the cells, of and cell and is associated with a inflammatory it has apparent that apoptosis of cardiac myocytes also occurs in a variety of important is a and form of cell death During apoptosis there is of and cell with of intracellular there is of and and of the cell into membrane which The process is The of during apoptosis energy and an the activation of which into that are of in form after on The also the expression of multiple which promote or the death (e.g. or and the expression of for (e.g. involved in cell destruction and coworkers have reported that NO produced by iNOS in macrophages can apoptosis of the macrophages and also of cells with the macrophages that release NO have also Szabolcs et al. used the Lewis to Wistar–Furth heterotopic heart transplantation model to investigate apoptosis occurs during cardiac allograft rejection and the time of apoptosis was related to the induction of iNOS found that significant apoptosis of macrophages and of cardiac myocytes occurred during cardiac allograft rejection. was identified by in of in cell by and also by the of on cardiac myocytes were by their for muscle and from the appearance of myocytes with and or In the rejecting allografts the of increased during days after transplantation and at day 5 there was a significant increase in endothelial cells and cardiac in to syngeneic grafts. The expression of iNOS mRNA protein and enzyme activity in the allografts in time and extent the apoptosis of cardiac myocytes. iNOS immunostaining of macrophages and cardiac myocytes also increased at days and was by positive immunostaining of cardiac muscle cells with an antibody to nitrotyrosine which is indicative of the formation of and The data are with but do not the that NO can apoptosis of macrophages and cardiac myocytes in this mechanisms such as interaction of cytotoxic T lymphocytes or cytokines with the protein or the activation of by released from T-lymphocytes also to apoptosis during rejection in rat heterotopic cardiac allograft for apoptosis and cardiac myocytes iNOS and nitrotyrosine Cardiac myocytes to inflammatory the rate of are by macrophages, which of their cellular macrophages for iNOS cardiac myocytes also iNOS which is after of cellular to is present in of myocyte damage and Other more recent studies have indicated that the drug but not apoptosis of isolated rat cardiac myocytes in vitro This effect was confirmed by of on and was inhibited by the addition to the media of reduced which NO released by the Treatment with (but not or reduced was also associated with the expression of the product which can be activated by to NO has been shown to inhibit synthesis and ribonucleotide and both NO and have been observed to cause damage and In to investigate apoptosis and increased expression of iNOS in human cardiac allograft rejection Szabolcs et al. ventricular endomyocardial biopsies from patients with cardiac allograft rejection grade with biopsies rejection grade using in end of and immunohistochemistry of both macrophages and of cardiac myocytes was observed in the rejecting allografts at a rate that was fold that observed in the of myocytes was present in both with inflammatory and in with immunostaining for iNOS was present in macrophages, cells and cardiac myocytes in the rejecting allografts Immunostaining for a of was found in the rejecting biopsies in iNOS protein expression was also present The data that apoptosis is a major form of cardiac myocyte death during human cardiac allograft rejection and that it occurs in association with the expression of iNOS and of indicative of the of cardiac allograft rejection for apoptosis and cardiac myocytes iNOS and nitrotyrosine to the experimental results in rat apoptosis of cardiac myocytes occurs to inflammatory in human allograft rejection. and cardiac myocytes for iNOS cardiac myocytes for nitrotyrosine to As previously, Lewis et al. found that the expression of iNOS mRNA in endomyocardial biopsies of human heart transplant recipients was inversely related to left ventricular performance, i.e. ejection fraction [15]. Since apoptosis produces the of cardiac myocytes, it is that myocyte by NO to the in ventricular function which is observed in cardiac allografts iNOS has also been demonstrated in ventricular tissues from animals with or myocardial infarction and from patients with In recent studies have demonstrated that apoptosis of cardiac myocytes occurred in at the of acute myocardial in Other studies have iNOS enzyme iNOS mRNA and protein in cardiac myocytes in myocardial tissues from patients with of have levels of cytokines and a myocardial inflammatory infiltrate recent have that apoptosis of cardiac myocytes occurs in ventricular tissue from patients with of advanced congestive heart failure apoptosis of heart muscle cells by iNOS to the of ventricular failure in these and other inflammatory of the of cardiac myocytes in association with iNOS expression in allograft rejection have several it the possibility that which inhibit iNOS enzyme activity or which with NO production by macrophages or be beneficial in reducing myocyte during allograft rejection or other such as In a rat model, aminoguanidine, a inhibitor of NO production by iNOS but which also has other prolonged survival and reduced the pathological changes in the grafts [10, 11]. In preliminary studies of our rat cardiac transplantation model, Yang et al. to the allograft recipients a which l-arginine into macrophages iNOS synthesis of NO and which also the release of TNF-α and other cytokines from macrophages In to treated control allografts, administration of was associated with of significant reduction of myocyte and of the myocardium the of apoptosis of cardiac myocytes in association with iNOS the possibility that one could such as of to the to myocyte it suggests that one cardiac myocyte death by inhibitors or of the such as which are involved in the of destruction of the cardiac myocytes This was in by