J Mol Cell Cardiol 28, 165–169 (1996)

Expression of Inducible Nitric OxideSynthase in Failing and Non-failingHuman HeartMartin Thoenes1, Ulrich Forstermann2, W. Ross Tracey3, Niels M. Bleese4,Andreas K. Nussler5, Hasso Scholz1 and Birgitt Stein1

1Abteilung Allgemeine Pharmakologie, Universitats-Krankenhaus Eppendorf, Universitat Hamburg,Martinistrabe 52, 20246 Hamburg, 2Pharmakologisches Institut, Universitat Mainz, ObereZahlbacher Strabe 67, 55101 Mainz, FRG, 3Abbott Laboratories, Abbott Park, Illinois, 60064, USA,4Abteilung Herz-u. Gefabchirurgie, Albertinen Krankenhaus, Suntelstrabe 11a, 22457 Hamburg, FRG,5Sektion Chirurgische Forschung, Universitat Ulm, Parkstrabe 11, 89073 Ulm, FRG

(Received 28 April 1995, accepted in revised form 24 July 1995)

M. T, U. F, W. R. T, N. M. B, A. K. N, H. S B. S. Expression ofInducible Nitric Oxide Synthase in Failing and Non-failing Human Heart. Journal of Molecular and Cellular Cardiology(1996) 28, 165–169. Recently, a significant activity of inducible nitric oxide synthase (iNOS) has been reportedin biopsies from failing hearts due to idiopathic dilated cardiomyopathy (IDC). Thus, a potential pathophysiologicalrole of iNOS in IDC has been stated. In order to investigate, whether iNOS expression is of pathophysiologicalrelevance in human heart failure, we measured iNOS protein expression and cGMP content in left ventricularmyocardium from non-failing and failing human hearts. Immunoblot analysis revealed iNOS protein expressionin four out of six failing hearts from septic patients, whereas no iNOS-protein expression was detected in eithernon-failing human hearts (n=6) or failing hearts due to IDC (n=9), ischemic heart disease (IHD, n=7), Beckermuscular dystrophy (BMD, n=2) and mitoxantrone-induced toxic cardiomyopathy (TCM, n=1). cGMP contentwas increased by 130% in septic hearts, whereas there was no cGMP increase in hearts with IDC, IHD and BMDcompared to non-failing hearts. We conclude, that the induction of iNOS may play a role in contractile dysfunctionobserved in septic shock, but is unlikely to be of major pathophysiological importance in end-stage heart failuredue to IDC, IHD, BMD and TCM. 1996 Academic Press Limited

K W: Nitric oxide; iNOS protein; cGMP; Heart failure; Sepsis.

et al., 1994). Furthermore, De Belder et al. (1993)Introductionfound a significant activity of inducible nitric oxidesynthase (iNOS) in right ventricular tissue fromHeart failure is the final manifestation of a het-

erogenous group of diseases where cellular al- patients with idiopathic dilated cardiomyopathy(IDC). The aim of the present study was to in-terations contribute to contractile dysfunction.

However, the mechanisms underlying patho- vestigate, whether the reported iNOS activity inIDC is due to an enhanced iNOS protein expression.physiology are still poorly understood. There is

evidence for a dysfunction of the -arginine-nitric In addition, we studied iNOS protein expression infailing hearts due to ischemic heart disease (IHD),oxide (NO) system in patients with heart failure,

since elevated plasma levels of nitrite/nitrate were Becker muscular dystrophy (BMD) and mi-toxanthrone-induced toxic cardiomyopathy (TCM).reported from patients with heart failure (Winlaw

Please address correspondence to: Martin Thoenes, Abteilung Allgemeine Pharmakologie, Universitats-Krankenhaus Eppendorf,Universitat Hamburg, Martinistrabe 52, 20246 Hamburg, FRG.

0022–2828/96/010165+05 $12.00/0 1996 Academic Press Limited165

M. Thoenes et al.166

From experimental studies in lipopolysaccharide A mouse monoclonal antibody raised against b-tubulin (Dianova, Hamburg, FRG) was choosen as(LPS)-treated animals and LPS- or cytokine-stim-

ulated cardiomyocytes there is evidence, that an an internal standard to evaluate equal loading ofprotein on each lane. Proteins were visualized byenhanced activity of iNOS within the myocardium

contributes to contractile dysfunction in septic enhanced chemiluminescence (Amersham, Buck-inghamshire, UK). The blocking experiment wasshock (Balligand et al, 1994; Stein et al., 1995).

Therefore, we also investigated iNOS expression in performed with ADP-sepharose-purified iNOS frommouse macrophages (RAW 264.7, supplied by W.hearts from patients with sepsis, where there are

elevated plasma levels of LPS and cytokines (De Ross Tracey) which was incubated overnight at4°C with the polyclonal antibody raised againstGroote et al., 1989). In order to obtain additional

information on alterations of the -arginine-NO- iNOS. cGMP content was measured by radio-immunoassay as described previously (Stein et al.,signal transduction system in heart failure, cGMP

content was measured in the same samples of 1993). In TCM, cGMP content was not measuredbecause of limited tissue available.myocardial tissue.

Materials and MethodsResults

Left ventricular myocardial tissue was taken fromexplanted human hearts with end-stage heart fail- Immunoblot analysisure due to IDC (n=9), IHD (n=7), BMD (n=2)and TCM (n=1) which were classified as NYHA Figure 1 depicts a representative immunoblot show-(New York Heart Association) stage III–IV. Left ing iNOS protein expression patterns in un-ventricular tissue from failing hearts of patients stimulated or stimulated macrophages and non-(n=5) with severe sepsis due to pneumonia (n= failing or failing human hearts. There is a prominent2) and peritonitis and mediastinitis (n=3) was band at about 130 kDa in stimulated macrophagesobtained during autopsy (8–58 h post mortem) and serving as positive control (RAW+) and also induring valve replacement due to bacterial endo- ventricular tissue from four out of six hearts ofcarditis (n=1). In order to exclude a significant patients with sepsis (SH), representing iNOS proteindegradation of proteins in the samples from septic expression in these samples. No band at abouthearts taken post mortem, control experiments were 130 kDa was detected in unstimulated macro-performed with samples from non-failing and failing phages serving as negative control (RAW−) as wellhearts taken as late as 64 h post mortem. Non- as in non-failing human hearts (NF, n=6) andfailing hearts (n=6) were obtained from multiorgan failing hearts due to IDC (n=9), IHD (n=7), BMDdonors whose hearts could not be used for surgical (n=2) and TCM (n=1). Interestingly, there is areasons or blood group incompatibility. Written small shift towards higher molecular weights ininformed consent according to the Helsinki De- septic human left ventricular tissue compared toclaration was obtained from all patients or the the macrophages. In Figure 2a cross-reactivity offamilies of prospective heart donors before car- the iNOS-antibody with human iNOS expressed indiectomy. Permission for these experiments was cytokine-stimulated human hepatocytes (HEP+) isobtained from the ethics committee of the Arzte demonstrated. Figure 2b shows blockade of theKammer Hamburg (Az 532/116/9.7.1991). Mouse antibody with ADP-sepharose-purified iNOS pro-macrophages (RAW 264.7, ATCC, Rockville, USA) tein. Compared to the unblocked antibody (SH−)were stimulated with c-interferon (3 U/ml, Gen- there is a clear reduction of the intensity of thezyme, Cambridge, USA) and lipopolysaccharide 130-kDa band in the septic hearts (SH+).from Escherichia coli (300 ng/ml, Sigma, Deisen-hofen, FRG) for 18 h. Samples were homogenized inLaemmli sample buffer with a microdismembrator(Braun, Melsungen; FRG). Two hundred micro- cGMP contentgrams of protein were electrophoretically separatedon a SDS-polyacrylamide slab gel (7.5%) according In ventricular tissue from patients with sepsis cGMP

content was increased to 230% (72±17 fmol/mgto Laemmli (1970) and transferred to nitrocelluloseas described previously (Neumann et al., 1988). ww, n=3) compared to non-failing human hearts

(100% or 31±6.9 fmol/mg ww, n=6). In myo-Proteins were detected using a rabbit polyclonalantibody raised against iNOS (Tracey et al., 1994). cardial tissue from patients with heart failure due

Expression of Inducible Nitric Oxide Synthase in Heart Failure 167

Figure 1 Immunoblot demonstrating patterns of inducible nitric oxide synthase (iNOS) protein expression in ventricularmyocardium from patients with heart failure. Left ventricular myocardium from patients with sepsis (SH), Beckermuscular dystrophy (BMD), ischemic heart disease (IHD), idiopathic dilated cardiomyopathy (IDC), mitoxantrone-induced toxic cardiomyopathy (TCM), non-failing hearts (NF) and unstimulated or stimulated mouse macrophages(RAW, −or+) for negative or positive control were loaded. There is a prominent band at about 130 kDa in stimulatedmouse macrophages as well as in septic hearts representing iNOS protein expression, whereas no band is detected atabout 130 kDa in unstimulated mouse macrophages, non-failing hearts and failing hearts due to BMD, IHD, IDC andTCM.

Figure 2 (a) Immunoblot demonstrating inducible nitric oxide synthase (iNOS) protein expression in stimulated mousemacrophages (RAW+, homogenate) and stimulated human hepatocytes (HEP+, cytosolic fraction). Two hundredmicrograms protein were loaded on each lane. The blot reveals cross-reactivity of the antibody between mouse andhuman iNOS. (b) Immunoblot demonstrating iNOS expression in septic hearts (SH). The right lane shows iNOS detectionwith the unblocked antibody. There is a prominent band at about 130 kDa (SH−). The left lane shows iNOS detectionafter the antibody had been incubated with ADP-sepharose-purified iNOS protein overnight at 4°C. There is a clearreduction of the intensity of the 130-kDa band (SH+).

M. Thoenes et al.168

post mortem. The band from the heart samples wasslightly shifted towards higher molecular weightsrelative to the stimulated mouse macrophagesamples. This is most probably due to species dif-ferences rather than tissue differences, since thereis also a small band shift between stimulated mousemacrophages and cytosolic fraction of stimulatedhuman hepatocytes (Fig. 2a). In addition, no bandshift was seen between cytosolic extract fromhuman hepatocytes and ventricular tissue fromseptic hearts (data not shown). De Belder et al.(1993) had reported significant iNOS activity in

80

0NF

*cG

MP

(fm

ol/m

g w

w)

IDCBMD IHD

60

40

20

SH

6 92 7 3

right ventricular biopsies from patients with IDC.Figure 3 cGMP content (pmol/mg ww) in left vent-In that study, atrial tissue from patients with IHDricular myocardium from non-failing and failing humanwere choosen as control. However, atrial and vent-hearts. The numbers in the columns denote the number

of hearts. Mean±... are given. cGMP content is higher ricular tissues differ in many morphological andin ventricular myocardium from patients with sepsis (SH) physiological aspects. Therefore, we used vent-than in non-failing hearts (NF), but was unchanged in ricular myocardium from non-failing hearts as thefailing hearts due to Becker muscular dystrophy (BMD),

control. In contrast to myocardial tissue from septicischemic heart disease (IHD) and idiopathic dilated car-patients, no iNOS band was detected in left vent-diomyopathy (IDC). ∗P<0·05 v NF.ricular myocardium from patients with IDC, IHD,BMD and TCM as well as in non-failing hearts. Thediscrepancies between the results from De Belder etto IDC, IHD and BMD cGMP content did not differal. (1993) and our study could be due to either thefrom that in non-failing human hearts (Fig. 3).lower sensitivity of immunoblot-detection comparedto the measurement of iNOS activity, or the differentsources of tissues used for analysis. In order toobtain information on the functional relevance ofDiscussioniNOS expression in human heart failure, we meas-ured cGMP content in the same samples (Fig. 3).The present study shows significant iNOS protein

expression in left ventricular myocardium from Since NO activates soluble guanylyl cyclase andincreases cGMP content, one would expect a cor-patients with sepsis. In order to demonstrate that the

130-kDa bands detected in septic hearts represent relation between iNOS expression and elevatedlevels of cGMP. Indeed, in ventricular myocardiumiNOS, the polyclonal antibody was incubated with

the purified iNOS protein before detection was per- from septic patients cGMP content was increasedby 130% compared to non-failing human hearts,formed. There was a significant reduction in the

intensity of the 130 kDa band, indicating iNOS whereas in ventricular myocardium from patientswith IDC, IHD and BMD cGMP content was un-protein expression in septic hearts (Fig. 2b). One

could argue, that iNOS protein might be altered in changed. This data supports our immunoblot find-ings showing iNOS protein expression in myocardialheart samples taken 8–58 h post mortem. Therefore,

control experiments were performed with myo- tissue from septic hearts and argues against anincreased iNOS protein expression in ventricularcardial samples from patients suffering from heart

failure or stroke (non-failing hearts) which were myocardium from patients with IDC, IHD and BMD.In the present study, the cellular localization oftaken 8–64 h post mortem and those obtained im-

mediately after transplantation. We did not find the induced enzyme cannot be exactly determined.However, it has been demonstrated, that LPS- andaltered protein expression patterns in ventricular

myocardium immediately frozen in liquid nitrogen cytokine-stimulated ventricular cardiomyocytes ex-press iNOS leading to an increase in cGMP content,after cardiectomy compared to the samples taken

post mortem. However, a degradation of iNOS pro- thereby mediating direct negative inotropic effectson cardiomyocytes (Balligand et al., 1994). Thus,tein in samples taken 48 h post mortem or later can

not be completely excluded, since Evans et al. (1994) iNOS protein expression in cardiomyocytes maycontribute to the contractile dysfunction observedfound that the half-life of iNOS-activity was 48 h.

This was also confirmed by our experiments where in septic shock, where plasma levels of endotoxinand cytokines are elevated. We conclude thatthe iNOS band was more prominent in myocardial

tissue taken 8 h than in samples taken 24 or 32 h an enhanced iNOS protein expression with a

Expression of Inducible Nitric Oxide Synthase in Heart Failure 169

Nitric oxide synthase activities in human myocardium.consecutive increase in cGMP could be patho-Lancet 341: 84–85.physiologically important in the cardio-

D G MA, M MA, D P, P MA,depression of septic shock, but is unlikely to play a W RP, 1989. Plasma tumor necrosis factor levelsmajor role in end-stage heart failure due to IDC, in patients with presumed sepsis. JAMA 262: 249–251.

E T, C A, C J, 1994. Inducible nitric-IHD, BMD and TCM.oxide-synthase mRNA is transiently expressed anddestroyed by a cycloheximide-sensitive process. Eur JBiochem 219: 563–569.

L UK, 1970. Cleavage of structural proteins dur-ing the assembly of the head of bacteriophage T4.AcknowledgementsNature 227: 680–685.

N J, S H, D V, S W, M-This study was supported by the Deutsche For- L, K P, 1988. Increase in myocardial Gi-

schungsgemeinschaft. Part of the data have been proteins in heart failure. Lancet October 22, 936–937.published as an abstract at the “First International S B, D A, M A, S W, S

H, 1993. Ca++-dependent constitutive nitric oxide syn-Conference on Biochemistry and Molecular Biologythase is not involved in the cyclic GMP-increasingof Nitric Oxide” (Los Angeles, USA, July 16–21,effects of carbachol in ventricular cardiomyocytes.

1994). J Pharmacol Exp Ther 266: 919–925.S B, F P, S W, S H, 1995. Long-

term treatment of cardiomyocytes with endotoxins andcytokines induces negative inotropic effects mediatedby nitric oxide. In: S. Moucada, M. Teelisch, R. Bude,

References E. A. Higgs (eds.) The Biology of Nitric Oxide, Physiologyand Clinical Aspects, Portland Press Proceedings Vol. 3,

B JL, U-L D, S WW, 85–89.P D, M TA, K M, T Z, T WR, X CH, K V, B J, PL CJ, D AJ, K RA, S TW, JS, F U, J RA, 1994. ImmunochemicalM T, 1994. Cytokine-inducible nitric oxide syn- detection of inducible NO synthase in human lung.thase (iNOS) expression in cardiac myocytes. J Biol Am J Physiol 266: L722–L727.Chem 269: 27580–27588. W DS, S GA, K AM, S CG,

D B AJ, R MW, W HJF, R PJ, S PM, MD PS, 1994. Increased nitricoxide production in heart failure. Lancet 344: 373–374.B CA, S E, M JF, M S, 1993.