Induction of Nitric Oxide Synthase in the Human Cardiac Allograft Is Associated With Contractile Dysfunction of the Left Ventricle
Background The mechanisms underlying cardiac contractile dysfunction after transplantation remain poorly defined. Previous work has revealed that inducible nitric oxide synthase (iNOS) is expressed in the rat heterotopic cardiac allograft during rejection; resultant overproduction of nitric oxide (NO) might cause cardiac contractile dysfunction via the negative inotropic and cytotoxic actions of NO. In this investigation, we tested the hypothesis that induction of iNOS may occur and be associated with cardiac allograft contractile dysfunction in humans.
Methods and Results We prospectively studied 16 patients in the first year after cardiac transplantation at the time of serial surveillance endomyocardial biopsy. Clinical data, the results of biopsy histology, and echocardiographic and Doppler evaluation of left ventricular systolic and diastolic function were recorded. Total RNA was extracted from biopsy specimens, and mRNA for β-actin, detected by reverse transcription–polymerase chain reaction (RT-PCR) using human specific primers, was used as a constitutive gene control; iNOS mRNA was similarly detected by RT-PCR using human specific primers. iNOS protein was detected in biopsy frozen sections by immunofluorescence. Myocardial cGMP was measured by radioimmunoassay, and serum nitrogen oxide levels (NOx=NO2+NO3) were measured by chemiluminescence. iNOS mRNA was detected in allograft myocardium at some point in each patient and in 59 of 123 biopsies (48%) overall. In individual patients, iNOS mRNA expression was episodic and time dependent; the frequency of expression was highest during the first 180 days after transplant (P=.0006). iNOS protein associated with iNOS mRNA was detected by immunofluorescence in cardiac myocytes. iNOS mRNA expression was not related to the ISHLT histological grade of rejection or to serum levels of NOx but was associated with increased levels of myocardial cGMP (P=.01) and with both systolic (P=.024) and diastolic (P=.006) left ventricular contractile dysfunction measured by echocardiography and Doppler.
Conclusions These data support a relation between iNOS mRNA expression and contractile dysfunction in the human cardiac allograft.
Cardiac transplantation is an effective treatment for end-stage heart failure, leading to improved functional status and 1- and 5-year survival >80% and >65%, respectively.1 After transplantation, episodic cardiac contractile dysfunction is frequently detected on routine echocardiography2 3 ; in certain patients, this dysfunction can markedly deteriorate and adversely affect outcome. The mechanisms underlying cardiac contractile dysfunction after transplantation are poorly understood. Despite the sparse cellular infiltrates and infrequent myocyte necrosis often encountered with histological rejection, contractile dysfunction is nevertheless usually attributed to cellular rejection. In many instances, however, severe allograft contractile dysfunction occurs in the presence of very minor histological change,4 suggesting that other poorly understood mechanisms may be responsible for allograft contractile dysfunction.
NO is a potent and widely distributed autacoid whose complex biological actions include the control of blood vessel wall function, neuronal transmission, and immune targeting. NO is enzymatically generated from its precursor l-arginine by three structurally distinct isoforms of NOS: endothelial, neuronal, and inducible NOS.5 These enzymes also differ functionally. The endothelial and neuronal NOS isoforms, referred to as constitutive, are regulated by intracellular calcium concentration and generate small amounts of NO in response to physical (eg, shear stress) and pharmacological stimulation.6 In contrast, the inducible isoform is not expressed in normal tissue but is transcriptionally upregulated in several tissues by LPS and cytokine stimulation, where it generates the sustained release of large amounts of NO.5 NO overproduction by iNOS in the systemic vascular bed is thought to underlie the vasodilatation and resistance to vasoconstrictors characteristic of sepsis.7 8 9 10 11 NO has also been shown to have profound effects on cardiac myocyte contractile function.12 13 14 15 After administration of LPS or cytokines, iNOS is induced in the myocardium of experimental animals8 16 ; in experimental sepsis models, myocardial iNOS induction is considered to contribute to impaired ventricular contractile performance via increased local NO generation.8 15 In humans, cytokine induction of myocardial iNOS with reduced cardiac contractility may also be present in sepsis17 18 and after the use of cytokines as antitumor therapy.19 20
Cardiac allograft rejection is associated with increased cytokine expression,21 22 and several cytokines have been shown to induce iNOS expression in myocardium.8 16 23 24 25 Acute rejection of the rat heterotopic cardiac allograft has recently been shown to be associated with induction of iNOS mRNA and protein in cardiac myocytes, cardiac microvascular endothelium, and infiltrating macrophages.26 These changes are associated with increased myocardial cGMP, the putative second messenger for the effects of NO.26 During rejection, NO formation is markedly increased in the rat heterotopic cardiac allograft, causing allograft protein nitrosylation27 and increased levels of NO degradation products in the systemic circulation.28 In the present study, we prospectively investigated whether induction of iNOS mRNA and protein occurs in human cardiac allografts and whether expression of iNOS mRNA is associated with cardiac contractile dysfunction. Our findings demonstrate induction of iNOS mRNA associated with protein expression, increased myocardial cGMP levels, and LV contractile dysfunction. We postulate that iNOS expression may contribute to contractile dysfunction of the human cardiac allograft by increased myocardial generation of NO, leading to increased cGMP levels, and that myocardial iNOS may represent a new target for therapeutic agents in the treatment of allograft contractile dysfunction.
The study protocol was approved by the Committee for the Protection of Human Subjects at Stanford University Medical Center, and written informed consent was obtained from all subjects before they were included in the study. Between January 1993 and January 1994, 16 consecutive adult patients undergoing cardiac transplantation who gave consent were recruited. Data were collected prospectively from all patients until June 1994. All patients were managed with a standard immunosuppressive regimen that included prophylactic antilymphocyte therapy (OKT3) during the early postoperative period and maintenance treatment with cyclosporin A, azathioprine, and tapering doses of prednisone. Episodes of moderate (grade 3) rejection, according to the ISHLT grading system,29 were treated with increased doses of corticosteroids, and in refractory cases, with antilymphocyte antibody. Adjunctive therapy with methotrexate was used in one patient for recurrent histological rejection. Full clinical data, posttransplant complications, and the results of routine laboratory investigations were documented. Samples of normal myocardium were obtained from 11 subjects at necropsy performed within 6 hours of sudden noncardiac death (intracranial hemorrhage 2, asphyxia 2, pulmonary embolus 1, carcinoma bronchus 1, butane overdose 1, murder 1, unknown 3).
Biopsy and Histological Grading
Surveillance right ventricular endomyocardial biopsy was performed via the right internal jugular vein with a modified Stanford-Caves-Schultz bioptome. Histological evaluation of rejection was performed by an experienced pathologist (M.E.B.) using the ISHLT grading system. One or two additional myocardial samples were immediately frozen in liquid nitrogen, after any adherent thrombus had been carefully removed, and stored at −80°C for subsequent experimental analysis. A 10-mL blood sample was drawn from the introducing sheath during the biopsy procedure, and serum was separated and stored at −80°C for measurement of serum nitrogen oxides (NOx=NO2−+NO3−). The routine surveillance biopsy regimen used at Stanford is weekly for the first 4 weeks after transplant, then every 2 weeks for 1 month, reducing gradually to once every 3 months, determined by the results of biopsy histology; biopsy was performed more frequently in the presence of histological rejection (ie, ISHLT grades 1 to 4).
Biopsies were homogenized in TRIzol reagent (Life Technologies Inc); total cellular RNA was extracted by the single-step acid guanidinium–thiocyanate-phenol-chloroform method.30
Reverse Transcription of mRNA and Generation of First-Strand cDNA
RNA (2 μg) was reverse transcribed to give cDNA in a final volume of 30 μL containing (final concentrations) Tris-HCl 10 mmol/L (pH 8.3), KCl 50 mmol/L, MgCl2 5 mmol/L, random hexamers 1.7 μmol/L, 0.5 mmol/L each of dATP, dTTP, dCTP, and dGTP, RNase inhibitor 0.5 U/μL, and Moloney murine leukemia virus reverse transcriptase 3.3 U/μL (Perkin-Elmer). The reaction was carried out at 42°C for 1 hour, followed by heat inactivation of the enzyme at 75°C for 10 minutes. cDNA was stored at −80°C.
Polymerase Chain Reaction
First-strand cDNA copies were amplified with Taq polymerase (Perkin-Elmer) and human-specific primers for β-actin cDNA31 (Table 1⇓), selected to control for adequate isolation of mRNA and conversion to cDNA or for iNOS cDNA based on the sequences reported for human hepatic and chondrocyte cDNAs32 33 and the iNOS chromosomal gene34 (Table 1⇓). Reactions were performed in a PTC 100-60 thermal cycler (MJ Research Inc) for 40 cycles. Annealing temperatures used were 50°C for β-actin, 60°C for iNOS (Clontech) and iNOS 1, and 63°C for iNOS 2 (Table 1⇓). Reactions were carried out in a final volume of 50 μL with 3 μL cDNA in (final concentrations) Tris-HCl 10 mmol/L (pH 8.3), KCl 50 mmol/L, 0.2 mmol/L each of dATP, dTTP, dCTP, and dGTP, 0.5 μmol/L primers, and Taq polymerase 0.02 U/μL, and MgCl2 concentration for each primer was set as shown in Table 1⇓.
Reproducibility of RT-PCR
Reproducibility of iNOS RT-PCR was examined in 10 biopsies (5 with iNOS mRNA expression and 5 without) using the second sample taken at the same surveillance biopsy. Because of the inherent nature of biopsy, the second sample is obtained from a different part of the interventricular septum.
PCR Product Resolution and Verification of Nucleotide Sequence
After amplification, 15 μL PCR products per lane was resolved on a 1.5% agarose gel containing ethidium bromide 0.5 μg/mL in 1× TBE buffer (Tris 0.45 mol/L, boric acid 0.45 mol/L, and EDTA 10 mmol/L). Bands were confirmed by two observers in blinded fashion after photography under UV fluorescence. PCR product band identity was determined in every case by Southern blot hybridization using a [γ-32P]ATP 5′ terminus-labeled (5′ DNA terminus labeling system; Gibco BRL) internal sequence probe for human specific iNOS (5′-CCCTCCTGTAGGCCCTC-3′). To probe for iNOS cDNA, the nylon membranes were prehybridized for 10 minutes in a solution containing 5× Denhardt’s solution (in wt/vol: Ficoll 0.1%, PVP 0.1%, BSA 0.1%), 0.5% SDS, and 5× SSPE (in mmol/L: NaCl 0.75, NaH2PO4 50, EDTA 5 [pH 7.4]) at 42°C and then hybridized in the same medium overnight at 42°C with the added radiolabeled DNA probe. After hybridization, the nylon membranes were washed twice in 2× SSC for 10 minutes at room temperature. The membranes were exposed to x-ray film (Kodak, X-OMAT AR) with intensifying screens for 16 hours at −70°C.
Exclusion of Genomic DNA Contamination of RNA Samples
To exclude contamination of RNA samples by genomic DNA (since the iNOS [Clontech] primers do not encompass an intron), 18 RNA samples with a positive iNOS PCR-amplified product were incubated at 37°C for 30 minutes with RNase A 1.8 μg/μL (Life Technologies Inc) before reverse transcription and PCR for iNOS and β-actin. Additionally, in 48 samples, PCR for iNOS was repeated with primers that encompassed two exon-intron junctions (iNOS 1, Table 1⇑), and the results of the two assays were compared in blinded fashion by two observers. The RNA content of each biopsy was also recorded to determine whether variations in RNA content influenced the result of the RT-PCR for iNOS mRNA.
Myocardial samples were fixed in paraformaldehyde (4% wt/vol) in PBS for 90 minutes, then dehydrated in an increasing gradient of sucrose in PBS, rapidly frozen by immersion in liquid nitrogen, and embedded in a 1:2 solution containing OCT compound (Miles Inc) and 20% sucrose in PBS. Sections (2 to 4 μm) were cut and thaw-mounted onto precleaned Superfrost Plus slides (Fisher Scientific Inc). After preincubation with 20% horse serum (Sigma Chemical Co) for 15 minutes, tissue cryostat sections were washed (2×10 minutes in PBS) and incubated at 4°C overnight with a rabbit anti-iNOS polyclonal antibody raised to a synthetic peptide derived from the C-terminal end of the mouse macrophage iNOS sequence (dilution 1:500; Affinity Bioreagents Inc) or rabbit IgG as negative control. Specific iNOS antibody binding was detected with a biotin-conjugated goat anti-rabbit IgG and strepavidin-conjugated Texas Red (Vector Laboratories, Inc). The slides were mounted and examined under an Olympus Vanox fluorescence microscope.
Specificity of the anti-iNOS antibody for human iNOS was determined by Western blot. Crude protein fractions (150 μg each) were separated on denaturing 7.5% SDS-PAGE gels, followed by blotting onto nitrocellulose filters. The blot was blocked with buffer (composition: 50 mmol/L Tris-HCl, pH 7.4, 0.15 mol/L NaCl, 2% BSA, 0.1% Tween-20) for 1 hour at room temperature, then incubated with the iNOS antibody (1:2000 dilution; Affinity Bioreagents Inc) for 1 hour at room temperature. The blot was washed six times with Tris-buffered saline (5 minutes each) and then incubated for 1 hour with anti-rabbit IgG antibody conjugated with HRP (Vector Laboratories Inc) at room temperature. The blot was washed six times (5 minutes each) with PBS, followed by detection of immunoreactive proteins by enhanced chemiluminescence (ECL, Amersham).
Myocardial cGMP Content
cGMP content was assayed in 10 biopsies (5 positive and 5 negative for iNOS mRNA expression). Biopsy tissue was homogenized in ice-cold 6% trichloroacetic acid (500 μL) and centrifuged (10 minutes, 4°C, 12 000 rpm); protein content in the pellet was measured with BioRad reagent (Bio-Rad Laboratories), and cGMP in the supernatant was measured by radioimmunoassay according to the manufacturer’s instructions (Cyclic GMP [125I] RIA kit, Du Pont).
Measurement of Serum Nitrogen Oxides
Serum nitrogen oxides (NO and one-electron oxidation products of NO [NOx]) were measured with a commercially available chemiluminescence detector (model 2108, Dasibi) after sample deproteinization using ethanol (1:3 vol/vol) and reduction in boiling acidic vanadium (III) chloride.35 By this technique, NO2− and NO3− are both quantitatively reduced to NO; NO is quantified by the chemiluminescence detector after reaction with ozone. Signals from the detector were analyzed by a computerized integrator and recorded as areas under the curve. Standard curves for NO2−/NO3− were linear over the range 100 pmol to 5 nmol, and serum was diluted to fall within this range. The assay was extensively validated against the Griess reaction, after reduction of serum NO3− to NO2− by nitrate reductase (data not shown).
Echocardiographic Assessment of LV Function
The echocardiographic examination included recordings of M-mode and two-dimensional images and pulsed-wave Doppler flow velocity signals across the mitral valve. All studies were performed with a Hewlett-Packard ultrasonograph using 2.5- or 3.5-MHz combined imaging and Doppler transducer. All studies were analyzed without prior knowledge of the endomyocardial biopsy findings or other clinical features of the patient. Standard recording techniques, recorded in detail elsewhere,36 37 were used. The following parameters of diastolic function were recorded: isovolumic ventricular relaxation time, peak early diastolic mitral flow velocity, and rate of peak early mitral flow deceleration, expressed as deceleration half-time as previously described.36 Intraobserver and interobserver correlation coefficients were .85 and .95, respectively, for measurements of isovolumic relaxation time, deceleration half-time, and peak early diastolic mitral flow velocity, as previously reported.37 Changes from baseline in any given patient were used as the index for a Doppler diagnosis of LV diastolic dysfunction; stable values are regarded as those that do not change beyond the threshold of spontaneous variations of the method, as previously reported.37 In these previous studies, a statistically significant change in isovolumic relaxation time or deceleration half-time has been determined to be one that exceeds 15%, and for peak early diastolic mitral flow velocity, one that exceeds 20%. In the present study, a significant decrease in isovolumic relaxation time or deceleration half-time formed the basis of a diagnosis of LV diastolic dysfunction. An increase in peak early diastolic mitral flow velocity in the presence of stable isovolumic relaxation time and deceleration half-time measurements was not regarded as indicative of diastolic dysfunction, owing to the propensity for this parameter to be influenced by heart rate. In the rare event of deceleration half-time and isovolumic relaxation time changing in opposite directions, an increase of ≥20% in peak early diastolic mitral flow velocity was used as an indicator of LV diastolic dysfunction.
Measurements of LV dimension were made in end systole and end diastole by the conventional American Society of Cardiography criteria. Percent fractional shortening derived from these measurements was compared in serial studies. Based on an intraobserver variation of 10%, systolic dysfunction was diagnosed when a decrease in percent fractional shortening >10% was observed between two serial studies. With these variables, LV diastolic and systolic function were described as categorical variables, ie, diastolic function normal or impaired, systolic function normal or impaired. Echocardiographic recordings were obtained on the day of surveillance biopsy in 74 of 123 cases (60%); systolic function was determined in all of these (60% of 123 biopsy data points). Diastolic parameters were measured in 58 (47%) of 123 biopsy data points; in 14 cases, diastolic parameters were not obtained because of equipment failure, and in 2 cases because of inadequate technical recordings. The frequency of abnormal echocardiographic LV function was analyzed in relation to iNOS mRNA expression as follows: systolic dysfunction alone, diastolic dysfunction alone, and either systolic or diastolic dysfunction (in this latter category, the 58 cases [47%] in which both parameters were recorded were analyzed).
Data are reported as group mean±SD. The significance of differences between groups was tested by unpaired Student’s t test or by contingency table analysis as appropriate. The association of multiple variables with iNOS expression was assessed by multiple regression analysis. A value of P<.05 was considered significant.
Sixteen male patients formed the study group (mean age, 48±14 years; range, 18 to 65 years). The indications for transplantation were ischemic heart disease in 10, dilated cardiomyopathy in 3, allograft coronary artery disease in 2, and acute graft failure in 1. One hundred thirty study biopsies (1 or 2 samples per biopsy) were obtained in these patients (4 to 12 serial samples per patient; range, 12 to 353 days after transplant). Of 130 biopsies, 123 (95%) were positive for β-actin mRNA expression (see below) and formed the group for further study. The donor organs for these 16 patients were obtained from 13 men and 3 women, age 28±10 years, ischemic time 186±48 minutes; 6 donors and 8 recipients were CMV positive, and CMV mismatch (negative recipient, positive donor status) was present in 1 patient. Complications encountered in this group were gastrointestinal CMV infection in 1 patient, pulmonary aspergillosis in 1, and jaundice of undetermined cause in 2. Two patients died: 1 of mucormycosis and 1 of chronic renal and multisystem failure. The ISHLT grades of rejection of this group of 123 biopsies were grade 0, 60 (49%); grade 1, 35 (28%); grade 2, 13 (11%); and grade 3, 15 (12%). The distribution of echocardiographic results within this group were systolic dysfunction, 23 of 74 (31%); diastolic dysfunction, 39 of 58 (67%); either systolic or diastolic dysfunction, 44 of 58 (76%); and normal function, 14 of 58 (24%). The mean age of subjects from whom normal myocardium was obtained at necropsy was 42±23 years.
PCR bands identifying the presence of β-actin mRNA were present in 123 of 130 transplant biopsies (95%); the 7 samples without a positive band (indicating inadequate RNA extraction or inadequate conversion of RNA to cDNA) were excluded from further analysis (Fig 1⇓). PCR bands indicating expression of iNOS mRNA were present in 59 of 123 biopsies (48%). iNOS mRNA was detected on at least one occasion in every patient, and expression was episodic as a function of time after transplantation (Fig 1⇓). iNOS mRNA expression was not detected in the 11 normal myocardial samples, which all showed positive β-actin bands (Fig 1⇓).
Reproducibility of RT-PCR
Reproducibility of the iNOS RT-PCR was confirmed in 10 transplant biopsies (5 expressing iNOS and 5 not) from RNA preparation through iNOS RT-PCR, showing agreement in every case. Since the two biopsies were taken from slightly different sites in the interventricular septum, this finding also suggests that expression of iNOS mRNA in transplant myocardium is homogeneous, at least within the interventricular septum where the biopsies were obtained.
RT-PCR Product Band Verification
PCR product band identity was confirmed in every case (59 of 59) by Southern blot hybridization using the 5′ end-labeled internal probe.
Exclusion of Genomic DNA Contamination of RNA Samples
Predigestion of RNA with RNase A before RT-PCR eliminated the PCR product band in 17 of 18 samples; in 1 sample, repeat PCR did not show a product band before or after RNase A digestion. iNOS RT-PCR was repeated using a primer set encompassing 2 introns (Table 1⇑, iNOS 1 primer set) showing agreement in samples with a positive band in 16 of 18 and in samples without a band in 24 of 30. These results suggest that significant genomic DNA contamination, which might influence the results of PCR, was not present. The mean amount of RNA isolated per biopsy was 5.5±4.7 μg in the 123 samples; RNA isolation was similar (P=.44) in the group without iNOS mRNA expression (5.8±5.6 μg, n=59) and in the group with iNOS mRNA expression (5.1±3.5 μg, n=64), suggesting that RNA isolation did not influence the results of RT-PCR.
iNOS Protein Detection by Immunofluorescence
The anti-iNOS antibody identified a 130-kD protein, representing rat iNOS, isolated from crude protein extracts of LPS-induced rat lung that was used as a positive control (lane 4, Fig 2B⇓, inset). The antibody also recognized a similar 130-kD protein isolated from the cytosol of cytokine-stimulated human colonic carcinoma cells, representing human iNOS (lanes 1 and 2, Fig 2B⇓, inset), kindly provided by Dr Paula Sherman.38 The antibody did not cross-react with endothelial NOS isolated from the particulate fraction of bovine aortic endothelial cells (lane 3, Fig 2B⇓, inset). With this antibody, iNOS protein was identified by immunostaining in 4 of 5 biopsies that showed iNOS mRNA expression by RT-PCR; no immunostaining for iNOS was detected in 3 of 4 biopsies without iNOS mRNA expression, and results were equivocal in the other 2 samples. The results obtained from representative biopsies with and without iNOS mRNA expression are shown in Fig 2⇓. iNOS protein immunostaining was detected in cardiac myocytes in all samples that expressed iNOS protein (myocytes labeled M, Fig 2A⇓ and 2B⇓). In 3 of 4 biopsies expressing iNOS protein, there was, in addition, staining of vascular smooth muscle cells in small intramyocardial vessels (indicated by arrows in Fig 2A⇓ and 2B⇓). These patterns of staining were not detected in samples that did not express iNOS mRNA (Fig 2C⇓ and 2D⇓).
Myocardial cGMP Content
Myocardial cGMP content was significantly increased (P=.01) in biopsies with iNOS mRNA expression compared with biopsies without iNOS mRNA expression (Fig 3⇓).
Serum Nitrogen Oxide Levels
Systemic venous serum samples from patients without iNOS mRNA expression showed NOx concentrations (48.7±35.4 μmol/L, n=38) similar to those from patients with iNOS mRNA expression (39.8±21.4 μmol/L, n=49; P=.15).
Relations of iNOS mRNA Expression to Clinical Variables
The relations of iNOS mRNA expression to clinical variables are shown in Table 2⇓ and in Figs 1b⇓ and 4⇓. Expression of iNOS mRNA was time dependent, the frequency of expression being highest in the first 180 days after transplant (Figs 1⇓ and 4⇓). iNOS mRNA expression was associated with higher prednisone dosage and with worse renal function (BUN and creatinine, Table 2⇓); however, multiple regression analysis showed that prednisone dosage, BUN, and creatinine were related to iNOS mRNA expression only through their relation to time after transplantation. The progressive steroid dose taper that is part of standard clinical care after transplantation and the time-dependent cumulative nephrotoxic effects of cyclosporin A, which progressively worsen renal function, appear to account for these associations.
Relation of iNOS mRNA Expression to Biopsy Histology
The mean ISHLT histological rejection grade was the same in the presence and absence of iNOS mRNA expression (Table 2⇑). The frequency of iNOS mRNA expression was similar in all grades of rejection (0 to 3, Fig 5⇓) and similar when rejection was analyzed as no rejection (grade 0) compared with rejection (grades 1 to 3, Fig 5⇓).
Relation of iNOS mRNA Expression to Echocardiographic LV Function
The relations of iNOS mRNA expression to LV function evaluated by echocardiography and Doppler are shown in Fig 6⇓. The frequencies of both abnormal systolic and diastolic functions were significantly increased in the group with iNOS mRNA expression. Echocardiographic LV function was completely normal (systolic and diastolic functions normal) in only 6% of the biopsies expressing iNOS mRNA. In contrast, echocardiographic LV function was completely normal in 46% of biopsies in which iNOS mRNA expression was absent.
Summary of Findings
The results of this study demonstrate, for the first time in humans, induction of iNOS mRNA and protein in the cardiac allograft associated with increased levels of myocardial cGMP and LV contractile dysfunction. The salient findings of the study are as follows. (1) iNOS mRNA expression was detected in ≈50% of routine surveillance biopsies after transplant; reproducibility of this finding was demonstrated in 10 biopsies, suggesting that iNOS mRNA expression is homogeneously distributed in transplant myocardium (at least within the interventricular septum). (2) iNOS mRNA expression occurred in every patient at some stage after transplant; expression was episodic after transplant and occurred most frequently during the first 180 days after transplant. (3) iNOS mRNA expression was associated with iNOS protein expression in cardiac myocytes and in the vascular smooth muscle cells of small intramyocardial vessels; iNOS expression was associated with increased myocardial levels of the intracellular second messenger of NO, cGMP. iNOS mRNA expression was not, however, associated with increased systemic serum levels of NOx (serum products of NO). (4) iNOS mRNA expression was unrelated to the presence of histological rejection. (5) iNOS mRNA expression was associated with increased frequencies of systolic and diastolic LV contractile dysfunction by echocardiography and Doppler. In the presence of iNOS mRNA expression, echocardiographic LV function was completely normal in only 6% of cases; conversely, LV function was completely normal in 46% of biopsies that did not express iNOS mRNA.
Relation of iNOS Expression to Cardiac Contractility
The mechanisms underlying contractile changes in the intact heart after cardiac transplantation are complex. Immune response–mediated vascular injury may influence cardiac myocyte function by altering vascular permeability and causing edema, by causing ischemia, or by altering the release of endothelium-derived substances, such as NO, that influence myocyte contractile function. Components of the immune system may also directly injure cardiac myocytes, leading to cell necrosis or dysfunction through the action of cytotoxic T lymphocytes, macrophages, and neutrophils; through pathways of humoral injury; and via the action of cytokines.2 Several studies have demonstrated that NO can modulate myocardial contractility. Exogenous NO and its second messenger, cGMP, reduce myocardial contractility.12 13 14 15 Cytokines reduce myocardial contractility acutely by stimulating constitutive NOS and increasing NO formation.13 Chronic exposure to cytokines has been shown to stimulate the transcriptional upregulation of iNOS and impair myocyte contractility through increased NO formation.8 12 13 16 In humans, myocardial dysfunction during sepsis and the therapeutic use of cytokines (as antitumor agents) may be mediated by induction of myocardial iNOS.17 18 19 20 Our data are consistent with the hypothesis that iNOS mRNA and protein expression occurs after cardiac transplantation and that this leads to increased myocardial NO generation within the allograft, causing increased myocardial cGMP levels and reduced myocardial contractility. The association of iNOS expression with diastolic dysfunction in our study is at variance with the results of a recent study39 showing improved diastolic distensibility during bicoronary infusion of sodium nitroprusside in patients with structurally normal hearts. However, it is important to recognize that the findings of Paulus et al39 are likely to reflect the physiological effects of small concentrations of NO on the heart at the concentrations of sodium nitroprusside infused. The effects of large amounts of NO on myocardial contractility, produced under pathological circumstances by the inducible isoform of NOS, are likely to be different. Furthermore, our data do not address the possibility that iNOS induction might adversely affect myocyte contractile function via multiple actions, such as the known inhibitory effects of NO on Fe2+-containing enzymes of the Krebs cycle and mitochondrial respiration5 ; nor have we addressed the possibility that iNOS expression might be associated with upregulation of other substances that have been shown to exert effects on myocyte contractility.40 Furthermore, the vascular expression of iNOS that we have described might result from a vascular injury whose effects on cardiac function are mediated by other mechanisms, such as the formation of interstitial edema or ischemia.2 It is also possible that iNOS induction could occur secondary to the reduction in myocardial contractility rather than contribute to it; preliminary data (by activity assay) showing induction of iNOS in inflammatory cardiac muscle disorders but not in ischemic heart disease with similar contractile dysfunction suggests that this is unlikely to be the case.41 Confirmation of the proposed direct effects of iNOS and NO on myocardial contractility in the human cardiac allograft awaits the development of specific inhibitors of iNOS. However, the use of inhibitors of iNOS in this situation could theoretically be detrimental; NO produced by graft-infiltrating macrophages has been shown to inhibit the cytotoxic T-lymphocyte response to alloantigen; inhibition of iNOS might thus promote T lymphocyte–mediated cytotoxicity.42
Relation of iNOS Expression to Biopsy Histology
Although biopsy histology has evolved as an invaluable tool for determining the level of immunosuppression to be used after transplant, it is nevertheless associated with certain biological and clinical inconsistencies. Histology is poorly correlated with markers of cytotoxic T-cell activation in the allograft43 and with cytokine gene expression.21 22 There is also a poor correlation between binding of immunoglobulin to the coronary microvasculature and biopsy histology.44 45 Last, there is limited association between the development of graft vascular disease, which is considered to have an immune basis, and antecedent histological rejection.46 The absence of a relation between iNOS expression and histology is therefore not without precedent. We speculate that histology represents one facet of the biology of cardiac allograft rejection, which coincides to an extent with other biological changes during the allograft immune response; histological change, however, may be present without evidence of concurrent activation of other immune components, eg, cytokines,21 22 and likewise, certain immune components may be activated without a concurrent change in histology.42 No single parameter can be regarded as sole mediator in the biology of rejection.
Comparison With Data Obtained in Animal Models
In a recent study examining iNOS expression in the rat heterotopic cardiac transplant model, iNOS mRNA and protein were identified in cardiac myocytes, in the microvasculature, and in infiltrating macrophages.26 These changes were associated with evidence of iNOS protein expression by activity assay and with increased myocardial cGMP levels. In earlier studies using the same model, rejection was accompanied by increased levels of serum oxides of nitrogen28 and evidence of graft protein nitrosylation,27 indicative of increased intragraft NO formation. It is not clear why the responses to cardiac allograft rejection differ somewhat in the rat compared with the human; however, the rat model is one of acute untreated rejection in a nonworking heart, in which induction of abnormal biochemical processes might be more marked, leading to larger increases in NO formation and thus NOx. Furthermore, iNOS induction in macrophages has been more readily demonstrable in several animal models than in humans32 47 ; infiltrating macrophages may be the major source of allograft NO generation in rats, leading to increases in serum NOx and graft protein nitrosylation. Last, in the present study it was not possible to regulate dietary nitrate intake during prolonged posttransplant follow-up, so that differences in nitrate intake may have caused alterations in serum NOx that would conceal smaller changes in endogenous NO production by the allograft. Differences in allograft NO generation related to myocardial iNOS expression may be more readily detected by sampling across the coronary vascular bed in future studies.
In animal models, steroid pretreatment generally inhibits the expression of iNOS induced by cytokines and LPS in several tissues.5 In apparent contrast, in the present study we found that iNOS mRNA expression was associated with higher steroid dosage. However, the association between iNOS expression and steroid dose was not an independent one; multiple regression analysis revealed that iNOS expression and steroid dose were correlated only through their independent associations with time after transplantation. iNOS expression itself exhibited marked time dependency after transplant: the highest frequency of expression was in the first 180 days after transplantation. Because high steroid doses are used early after transplantation and subsequently tapered, there is an apparent but spurious association between iNOS expression and higher steroid dosage. On the basis of these data, we cannot confirm or exclude a relation between steroid dosage and iNOS expression in human cardiac transplantation.
A limitation of this study, in common with any clinical study involving limited amounts of tissue, is the inability to measure every parameter in every biopsy. The amounts of mRNA and protein present in such small specimens (5 mg) precluded application of quantitative methods such as Northern and Western blotting. We adopted the approach of using iNOS RT-PCR as a screening technique to detect iNOS gene transcription in small samples of allograft myocardium, after demonstrating that iNOS mRNA is homogeneously expressed in myocardium. We subsequently performed other tissue measurements and categorized samples according to iNOS mRNA expression by RT-PCR. A further limitation in the interpretation of our results is that the biopsy samples assessed in this study did not include any samples taken at the time of a clinically severe reduction in myocardial contractile function. Accordingly, we cannot be certain that our extrapolations relating iNOS induction to graft contractile dysfunction apply to the development of cardiogenic shock occurring as a consequence of immune injury after transplantation.
We have shown episodic, time-dependent expression of iNOS mRNA in the human cardiac allograft associated with iNOS protein expression, elevated myocardial cGMP concentrations, and ventricular contractile dysfunction. We postulate that iNOS expression and contractile dysfunction are causally related via the local effects of NO and cGMP on cardiac myocyte function. Future studies directed at an understanding of the factors responsible for iNOS expression in the allograft and the long-term consequences of iNOS expression will be of interest.
Selected Abbreviations and Acronyms
|iNOS||=||inducible nitric oxide synthase|
|ISHLT||=||International Society for Heart and Lung Transplantation|
|NOS||=||nitric oxide synthase|
|PCR||=||polymerase chain reaction|
|RT-PCR||=||reverse transcription PCR|
Dr Lewis is the recipient of a British Heart Foundation International Fellowship. Dr Tsao is the recipient of a National Service Research Award (1F32-HL-08779). Dr Rickenbacher is the recipient of a grant from the Margarete and Walther Lichtenstein Foundation. Dr Haywood is the recipient of a grant from the Wessex Cardiothoracic Center Research Trust. Dr von der Leyen is the recipient of an award from the Deutsche Forschungsgemeinschaft (Le 567/3-1 and 3-2). Dr Cooke is a recipient of the Vascular Academic Award from the National Heart, Lung, and Blood Institute. Dr Valantine is supported by NIH Clinical Investigator grant 5K08-HL-02477.
Presented in part at the 67th Scientific Sessions of the American Heart Association, Dallas, Tex, November 14-17, 1994.
- Received April 3, 1995.
- Revision received September 12, 1995.
- Accepted September 25, 1995.
- Copyright © 1996 by American Heart Association
The Registry of the International Society for Heart and Lung Transplantation: Tenth Official Report. J Heart Lung Transplant. 1993;12:543-548.
Barry WH. Mechanisms of immune-mediated myocyte injury. Circulation. 1994;89:2421-2432.
Moncada S, Palmer RMJ, Higgs EA. Nitric oxide: physiology, pathophysiology, and pharmacology. Physiol Rev. 1991;43:109-142.
Cooke JP, Rossitch E Jr, Andon NA, Loscalzo J, Dzau VJ. Flow activates an endothelial potassium channel to release an endogenous nitrovasodilator. J Clin Invest. 1991;88:1663-1671.
Knowles RG, Merrett M, Salter M, Moncada S. Differential induction of brain, lung and liver nitric oxide synthase by endotoxin in the rat. Biochem J. 1990;270:833-836.
Nava E, Palmer RMJ, Moncada S. The role of nitric oxide in endotoxic shock: effects of NG-monomethyl-L-arginine. J Cardiovasc Pharmacol. 1992;20(suppl):S132-S134.
Balligand JL, Ungureanu D, Kelly RA, Kobzik L, Pimental D, Michel T, Smith T. Abnormal contractile function due to induction of nitric oxide synthase in rat cardiac myocytes follows exposure to activated macrophage-conditioned medium. J Clin Invest. 1993;91:2314-2319.
Finkel MS, Odis CV, Jacob TD, Watkins SC, Hattler BG, Simmons RL. Negative inotropic effects of cytokines on the heart mediated by nitric oxide. Science. 1992;257:387-389.
Brady AJB, Warren JB, Poole-Wilson PA, Williams TJ, Harding SE. Nitric oxide attenuates cardiac myocyte contraction. Am J Physiol. 1993;265:H176-H182.
Brady AJB, Poole-Wilson PA, Harding SE, Warren JB. Nitric oxide production within cardiac myocytes reduces their contractility in endotoxemia. Am J Physiol. 1992;263:H1963-H1966.
Nora R, Abrams JS, Tait NS, Hiponia DJ, Silverman HJ. Myocardial toxic effects during recombinant interleukin-2 therapy. J Natl Cancer Inst. 1989;81:59-63.
Kinugawa K, Takahashi T, Kohmoto O, Yao A, Aoyagi T, Momomura S, Hirata Y, Serizawa T. Nitric oxide–mediated effects of interleukin-6 on [Ca2+]i and cell contraction in cultured chick ventricular myocytes. Circ Res. 1994;75:285-295.
Balligand JL, Ungureanu D, Pimental D, Kelly RA, Smith TW, Michel T. Detection of transcripts for a cytokine-inducible nitric oxide synthase isoform in adult rat ventricular myocytes. Circulation. 1993;88(pt 2):II-1-II-384.
Yang X, Chowhury N, Cai B, Marboe S, Sciacca RR, Michler RE, Cannon PJ. Induction of myocardial nitric oxide synthase by cardiac allograft rejection. J Clin Invest. 1994;94:714-721.
Lancaster JR, Langrehr JM, Bergonia HA, Murase N, Simmons RL, Hoffman RA. EPR detection of heme and nonheme iron-containing protein nitrosylation by nitric oxide during rejection of the rat heart allograft. J Biol Chem. 1992;267:10994-10998.
Langrehr JM, Murase N, Markus PM, Cai X, Neuhaus P, Schraut W, Simmons RL, Hoffman RA. Nitric oxide production in host-versus-graft and graft-versus-host reactions in the rat. J Clin Invest. 1992;90:679-683.
Billingham ME, Carey NRB, Hammond ME, Kemnitz J, Marboe C, McAllister HA. A working formulation for the standardization of nomenclature in the diagnosis of heart and lung rejection: Heart Rejection Study Group. J Heart Lung Transplant. 1990;9:587-592.
Shi Y, Hutchinson H, Hall DJ, Zalewski A. Downregulation of c-myc expression by antisense oligonucleotides inhibits proliferation of human smooth muscle cells. Circulation. 1993;88:1190-1195.
Geller DA, Lowenstein CJ, Shapiro RA, Nussler AK, Di Silvio M, Wang SC, Nakayama DK, Simmons RL, Snyder SH, Billiar TR. Molecular cloning and expression of inducible nitric oxide synthase from human hepatocytes. Proc Natl Acad Sci U S A. 1993;90:3491-3495.
Charles IG, Palmer RMJ, Hickery MS, Bayliss MT, Chubb AP, Hall VS, Moss DW, Moncada S. Cloning, characterization and expression of a cDNA encoding an inducible nitric oxide synthase from the human chondrocyte. Proc Natl Acad Sci U S A. 1993;90:11419-11423.
Chartrain NA, Geller DA, Koty PP, Sitrin NF, Nussler AK, Hoffman EP, Billiar TR, Hutchinson NI, Mudgett JS. Molecular cloning, structure and chromosomal localization of the human inducible nitric oxide synthase gene. J Biol Chem. 1994;269:6765-6772.
Tsao PS, McEvoy LM, Drexler H, Butcher EC, Cooke JP. Enhanced endothelial adhesiveness in hypercholesterolemia is attenuated by l-arginine. Circulation. 1994;89:2176-2182.
Paulus WJ, Vantrimpont PJ, Shah AM. Acute effects of nitric oxide on left ventricular relaxation and diastolic distensibility in humans: assessment by bicoronary sodium nitroprusside infusion. Circulation. 1994;89:2070-2078.
Shah AM, Mebazaa A, Wetzel RC, Lakatta EG. Novel cardiac myofilament desensitizing factor released by endocardial and vascular endothelial cells. Circulation. 1994;89:2492-2497.
De Belder A, Radomski M, Why H, Richardson P, Martin J, Moncada S. Expression of an inducible NO synthase in heart tissue from patients with inflammatory heart disease. J Am Coll Cardiol. 1994;(suppl):263A. Abstract.
Miller LW, Wesp A, Jennison SH, Graham MA, Martin TW, McBride LR, Pennington DG, Peigh P. Vascular rejection in heart transplant recipients. J Heart Lung Transplant. 1993;12(pt 2):S147-S153.
Hibbs JB, Westenfelder C, Taintor R, Vavrin Z, Kablitz C, Baranowski RL, Ward JH, Menlove RL, McMerry MP, Kushner JP, Baranowski WE. Evidence for cytokine-induced nitric oxide synthesis from L-arginine in patients receiving interleukin-2 therapy. J Clin Invest. 1992;89:867-877.