OUP user menu

Positive inotropic effects of NO donors in isolated guinea-pig and human cardiomyocytes independent of NO species and cyclic nucleotides

David Sarkar, Patrick Vallance, Charles Amirmansour, Sian E. Harding
DOI: http://dx.doi.org/10.1016/S0008-6363(00)00202-9 430-439 First published online: 1 December 2000


Objective: To characterise the inotropic response of isolated myocytes to a range of structurally unrelated NO donors and to assess the role of NO release kinetics, NO species and cyclic nucleotides in mediating the observed changes. Method: Guinea-pig (GP) and human myocytes were prepared by enzymatic digestion. Paced contractile amplitude was recorded at 37°C. NO release was measured by reduction of oxyhaemoglobin and using an NO electrode. Cyclic nucleotides were measured using a tritium labelled competitive binding assay. Results: The NO donors S-nitrosoglutathione (GSNO) and diethylamine/NO (DEA/NO) produced positive inotropic effects in GP myocytes at (10−5 M) (25 and 111% increases of contraction amplitude).The response to GSNO was significantly enhanced in the presence of a low concentration of isoprenaline (3×10−10 M). Positive inotropy was observed with a range of both thiol and non-thiol donors, amongst which a fast rate of NO release was associated with positive inotropy. The response to GSNO was abolished by the free NO scavenger oxyhaemoglobin, but not by ODQ (soluble guanylyl cyclase [sGC] inhibitor), Rp-cAMPS (protein kinase A inhibitor) or thapsigargin (sarcoplasmic reticulum Ca2+ uptake blocker). Direct measurement of cyclic nucleotides showed a rise in cGMP but not cAMP. Human ventricular myocytes showed a significant increase of contraction with GSNO (48±15.8%, n = 7, P<0.05) in the presence of isoprenaline and a marked response to DEA/NO alone. Conclusions: Isolated GP and human myocytes show a positive inotropic effect with certain NO donors. This is independent of sGC and cAMP. The rate of NO release from donors appears important in mediating the effect.

  • Myocytes
  • Nitric oxide
  • Contractile function

Time for primary review 35 days.

1 Introduction

Nitric oxide (NO) plays an important role in cardiovascular physiology and pathophysiology. Its effects on the modulation of vascular tone have been well documented [1] but its action on myocardial performance remains controversial.

There is extensive evidence for NO generation within the heart [2,3], though the effect of endogenously produced NO on contractile function remains contentious. Inhibition of endogenous NO has no effect on isolated myocyte contractile function in guinea-pig [4], rat [5], dog [6], or the failing human heart [7]. This suggests an absence of a reversible tonic NO effect. Some reports indicate that the response to β-adrenergic stimulation of rat cells and failing human hearts in vivo is enhanced in the presence of a NOS inhibitor [8,9], indicating that NO may modulate autonomic influences on the heart. However, this modulatory effect has not been a universal finding [7,10]. Induction of inducible nitric oxide synthase (iNOS) by endotoxaemia [4] or cytokines depresses contractility which recovers with NOS inhibition. Cardiac iNOS has been detected in a range of clinical situations and may account for the depression of myocardial contractility seen in septic shock [11,12], cardiomyopathy [13,14] and allograft rejection [15]. The prevailing view is that endogenously produced NO from constitutive or induced nitric oxide synthase has either no effect or is a weak negative inotrope.

Reports of the effect of exogenous NO on contractile function are contradictory. Initial studies of the effect of NO on cardiac muscle or myocytes demonstrated a modest depression of contractility with sodium nitroprusside (SNP) or dissolved NO gas, reducing contraction by 20–30% [5,16]. However, subsequent reports suggest that low doses of certain NO donors potentiate rather than depress adrenergic stimulation of rat myocytes [17]. Furthermore, human and feline papillary muscle show a positive response to glyceryl trinitrate (GTN) [18], SNP and 3-morpholinosydnonimine (SIN 1) [19] whilst a biphasic response has been observed with S-nitroso-N-acetylpenicillamine (SNAP) [19]. In patients undergoing cardiopulmonary bypass a variable inotropic response to SNP was observed, the direction of which was related to the presence or absence of preoperative β-blockade [20]. It is unclear why endogenous and exogenous NO might produce differing results or why there are so many contradictory reports of the effects of exogenous NO on myocyte function.

The aims of the present study were (i) to characterise the response of isolated guinea-pig myocytes to a range of structurally unrelated NO donors and to test the hypothesis that inotropic responses are dependent upon either the rate of NO release or species of NO generated; (ii) to define the role of cGMP and cAMP in mediating the changes seen and (iii) to investigate to what extent functional changes in guinea-pig cardiomyocytes might also occur in isolated human cells.

The isolated myocyte system was chosen to remove any potential paracrine effects mediated through endothelial or other non-myocytic cells within the intact cardiac preparation.

2 Methods

2.1 Preparation of isolated myocytes

The studies of human tissues and cells conformed to the principles outlined in the declaration of Helsinki whilst animal investigations conformed with the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH publication No. 85-23, revised 1996).

The myocyte isolation protocol has been previously published [16]. Male Dunkin–Hartley guinea-pigs were injected with intraperitoneal (i.p.) heparin (1000 units/kg) 15 min prior to sacrifice by cervical dislocation. The beating heart was rapidly excised and immersed in ice-cold Krebs–Henseleit (KH) solution (composition mM: NaCl 119, CaCl2 1.0, KCl 4.7, MgSO4 0.94, KH2PO4 1.2, NaHCO3 25 and glucose 11.5) gassed with 95% O2– 5% CO2. The heart was perfused retrogradely in a Langendorff mode, at a flow-rate of 8 ml/g of tissue with low calcium solution (composition mM: NaCl 120, KCL 5.4, MgSO4 5.0, Ca2+ 0.012–0.015, pyruvate 5.0, taurine 20, HEPES 10, glucose 20, nitrilotriacetic acid 5.0 and gassed with 100% oxygen at a pH of 6.95). The 5 min low calcium perfusion was followed by the first enzyme digestion step using 4 Uml−1 of protease (Sigma type XXIV protease) for 1 min. This was followed by collagenase (0.3 mg/ml, type II Worthington, New Jersey, USA.) and hyaluronidase (0.3 mg/ml type I Sigma). Post perfusion the left ventricle was removed and manually divided into small pieces. These were subjected to a further 5 min collagenase digestion. The resultant cell suspension was filtered through a 300-μm gauze, centrifuged at 1000 rpm for 60 s and the pellet washed and resuspended. The undigested tissue underwent a repeat collagenase digestion. Generally the yield of viable myocytes was greatest after the repeat collagenase digestion and these were the cells used in the majority of experiments.

Samples of failing human myocardium were obtained from explanted hearts at transplantation. In addition, small biopsies of the LV were obtained from patients undergoing routine mitral valve replacement in the absence of coronary disease. Our protocol for human myocyte isolation has been published previously [7].

2.2 Measurement of contractile function

Cells were superfused at 2 ml/min with Krebs–Henseleit solution at 37°C on the stage of a Zeiss IM inverted microscope. Contraction was induced by a bipolar stimulator via platinum electrodes on either side of the bath. Stimulation frequency was 0.5 Hz for guinea-pig and 0.2 Hz for human. Contraction amplitude was recorded using a video edge detection system running at 50 or 100 Hz with a spatial resolution of 1 in 256 or 512 [21].

2.3 Measurement of spontaneous NO release

Spontaneous release of NO from a range of donors was determined in an acellular system by measurement of the reduction of oxyhaemoglobin to methaemoglobin [22]. Oxyhaemoglobin (5×10−6 M) dissolved in KH solution with 1 mM Ca2+, at 37°C, was placed in a quartz cuvette and the NO donor added. Using a dual wavelength spectrophotometer (Shimadzu UV-3000) the change in absorbance at 401 and 411 nm was continuously measured. Liberated NO converts oxy- to methaemoglobin causing an increase in absorption at 401 and little change at 411 nm. NO production was calculated using the equation

Embedded Image

To investigate the effect of myocytes on the rate of NO donor decomposition parallel measurements of NO release were performed in the absence and presence of 500 000 guinea-pig myocytes using an NO electrode (iso-NO with 2 mm shielded sensor; World Precision Instruments, Sarasota, FL, USA). Incubations were carried out in HEPES-buffered Krebs–Henseleit solution at pH 7.4 and 37°C (composition mM: NaCl 118, KCl 4.8, KH2PO4 1.2, MgSO4 1.2, CaCl2 1.0, glucose 11.5 and HEPES 25). The NO donors were added from aqueous stock solution by means of a Hamilton syringe. Changes in current output were recorded continuously and processed using the duo-18tm data acquisition software (version 1.1; World Precision Instruments). NO release was quantified by comparison with a standard curve constructed by addition of known amounts of NaNO2 to a solution of potassium iodide in sulphuric acid.

2.4 Measurement of cAMP

One hundred thousand viable rod-shaped myocytes were suspended in 1 ml of buffer solution (composition mM: NaCl 120, KCL 5.4, MgSO4 5, pyruvate 5, glucose 20, taurine 20, HEPES 10, CaCl2 0.2). No attempt was made to remove non viable myocytes. Cells were incubated as a suspension at 37°C for 5 min followed by a further 5 min under test conditions. The cells were separated by refrigerated centrifugation for 10 s at 15 000 rpm. The supernatant was discarded and the cells resuspended in 1 ml of ice cold acidified ethanol containing isobutyl-1-methylxanthine (IBMX 10−4 M). Following cell lysis the samples were dried at 57°C under nitrogen to prevent nucleotide oxidation. Samples were then resuspended in 100 μl of sample buffer and assayed using a tritium labelled competitive binding assay (TRK 432, Amersham, Bucks., UK). Each experimental condition was performed in triplicate for each myocyte preparation. The results were expressed as a percentage of the basal level for each preparation.

2.5 Measurement of cGMP

Measurements were made on 200 000 isolated myocytes preincubated with IBMX 10−4 M and Zaprinast 10−5 M in order to optimise the yield of cGMP. Cells were exposed to experimental conditions (GSNO 10−5 M, DEA/NO 10−5 M or DEA/NO 10−5 M and YC-1 10−4 M) for 3 min followed by centrifugation and lysis as outlined above [1,2,4]. oxadiazolo[4,3-a]quinoxalin-1-one (ODQ 10−5 M) was added to the preincubation mixture of IBMX and Zaprinast to assess its inhibitory effect on the response to GSNO. Cyclic GMP levels were assayed using a tritiated competitive binding assay (TRK 500, Amersham).

2.6 Experimental protocol

In initial studies, concentration response curves to isoprenaline (10−10–10−7 M) were constructed. After reaching a maximal effect or inducing spontaneous contractile activity, the cells were washed for 15 min and the concentration–response curve was repeated in the presence or absence of l-NAME (10−4 M).

Separate cells were treated with a continuous low concentration of isoprenaline and contraction was assessed with co-infusion of increasing concentrations of an NO donor. The following donors were studied over the concentration range 10−6–10−4 M, SNP, GTN, S-nitrosoglutathione (GSNO), SNAP, diethylamine nitric oxide (DEA/NO), detanonoate and papanonoate. The steady state contraction amplitude was recorded after 5 min exposure to each concentration of donor drug.

The mechanism of the response to GSNO, in the presence of low dose isoprenaline, was studied by superfusing contracting myocytes for 10 min with various inhibitors including ODQ [23] (soluble guanylyl cyclase (sGC) inhibitor), thapsigargin [24] (sarcoplasmic reticulum Ca2+ uptake blocker) and oxyhaemoglobin (NO scavenger). The effect of ODQ on the response to DEA/NO alone was also tested.

To examine the role of protein kinase A, cells were stimulated with DEA/NO in the absence of isoprenaline. Having induced a response the cells were washed and treated with the inhibitor Rp-cAMPS (2×10−4 M) by recirculation for 40 min, then rechallenged by the addition of DEA/NO to the recirculation medium.

Human cells were stimulated with isoprenaline and after completion of a concentration–response curve cells were challenged with GSNO in the presence of a low concentration of isoprenaline (3×10−10–3×10−9 M) or by DEA/NO alone.

2.7 Statistical analysis

Values are expressed as means±SEM except where stated. Statistical differences between basal contraction and that seen under experimental conditions were tested using Student's two-tailed t-test for paired data with the exception of the comparison of the effect of low dose isoprenaline on the response to GSNO which was made with an unpaired t-test. The n value refers to the number of cells used in a particular experiment with each coming from a separate animal or human preparation unless otherwise stated.

2.8 Drug source and preparation

GSNO was synthesised by Dr. D. Madge, Medicinal Chemistry, Wolfson Institute for Biomedical Research UCL, London, UK. Papanonoate, detanonoate and ODQ were from Alexis (Nottingham, UK). DEA/NO, and SNAP were purchased from RBI (Natick, MA, USA). Carboxy-PTIO was from Calbiochem (La Jolla, CA, USA). All other reagents were obtained from Sigma (Poole, Dorset, UK).

Stock solutions of NO donors were prepared in deionised water, immediately prior to use, and kept on ice protected from light exposure. ODQ, IBMX and Zaprinast were dissolved to 10−2 M in DMSO. Isoprenaline stock dilutions were made in KH solution containing ascorbic acid (10−3 M) and diluted 1000 fold in the superfusing solution.

3 Results

3.1 Effects of endogenous NO

Isoprenaline 10−10–10−7 M induced a concentration-dependent increase in contraction amplitude which was reproducible when repeated twice in a single study. The concentration response to isoprenaline was not significantly different when repeated in the absence or presence of l-NAME (10−4 M) (EC50 12.9±4.5 nM, with l-NAME 17.6±7.4 nM). The maximum contraction amplitude was 9.6±0.7% shortening without l-NAME and 8.8±0.8% with l-NAME, n = 7. l-NAME had no effect on basal contraction amplitude in the absence of isoprenaline.

3.2 Effect of low dose isoprenaline on response to NO donors

To determine the effects of a background concentration of isoprenaline on the response to NO donors, cells were stimulated with increasing concentrations of GSNO in the presence or absence of isoprenaline (3×10−10 M). Basal contraction amplitudes in the presence and absence of low dose isoprenaline were not significantly different (2.86±0.23 vs. 3.09±0.45% of resting cell length). GSNO at 10−5 M and 3×10−5 M produced a significant increase in contraction amplitude in the absence of β-adrenergic stimulation (25.4±7.5%, n = 10, P<0.01: and 29.5±12.4%, n = 9, P<0.05, respectively). In the presence of isoprenaline the magnitude of the positive inotropic effect was greater (10−5 M GSNO, 70.9±8.3%, n = 7, P<0.01: 3×10−5 M GSNO, 92.5±10.2%, n = 17, P<0.001). Direct comparison of the effect of GSNO at both concentrations (10−5 M and 3×10−5 M) revealed a significantly larger inotropic response in the presence of isoprenaline (10−5 M, P<0.01 and 3×10−5 M, P<0.001). In view of this enhancement a low background concentration of isoprenaline (3×10−10 M) was used for subsequent experiments involving GSNO.

The potent NO donor DEA/NO was capable of inducing a large positive inotropic effect without the need for background isoprenaline. DEA/NO (10−5 M) alone raised contraction amplitude from 3.55±0.40 to 7.33±0.76%. The average change of contraction amplitude was 112±14% (n = 8, P<0.0005).

3.3 Effect of exogenous NO donors

In subsequent studies cells were exposed to a range of different NO donors at single or cumulative concentrations in the presence of isoprenaline (3×10−10 M). Changes of contraction amplitude in the presence of an NO donor (10−5 M) were expressed relative to the control value of isoprenaline alone (Fig. 1a). At no concentration tested did the clinically used NO donors GTN and SNP (10−6–10−4 M) significantly reduce contraction amplitude. At 10−5 M the reductions were 1.5 and 12%, respectively. Detanonoate produced a small but nonsignificant increase of contraction. In contrast GSNO, SNAP, papanonoate and DEA/NO (10−5 M) produced large increases in contraction amplitude (GSNO: 70.9±8.3%, P<0.001; SNAP: 111±21.%, P<0.005; papanonoate: 91.4±6.3%, P<0.001, DEA/NO 168±59%, P<0.005). The increase in amplitude was of rapid onset, was sustained for the duration of exposure to the drug and reversed with removal of the NO donor (Fig. 1b). The positive inotropic response to GSNO generally occurred at 10−5 M. Further increase of NO donor concentration did not yield an additional increase in the inotropic response. Between 10−6 M and 10−4 M there was no evidence of a biphasic response. Glutathione alone (10−6–10−4 M) did not cause a significant increase in contraction while the same cells responded to GSNO (10−5 M) by a doubling of contraction amplitude. Similarly the cells which showed no significant change with GTN (10−6–10−4 M) when subsequently challenged with SNAP (10−5 M) produced a 94±8% increase in contraction amplitude (P<0.005), confirming these cells were capable of mounting a positive inotropic response.

Fig. 1

(a) Effects of GTN (n = 4), SNP (n = 5), GSNO (n = 7), SNAP (n = 7), papanonoate (n = 6), DEA/NO (n = 5) and detanonoate (n = 8) at 10−5 M, on isolated guinea-pig myocytes. Contractile responses are displayed as a percentage above the control value of isoprenaline (3×10−10 M) alone. Each bar represents the mean±SEM. **, P<0.005 and ***, P<0.001. (b) Representative tracing of contraction amplitude of an isolated guinea-pig myocyte exposed to GSNO 10−5 M. The amplitude increases by 35% (in the absence of isoprenaline) and this effect is fully reversed once the GSNO is washed out.

3.4 Oxyhaemoglogin

Seven cells from five preparations were serially exposed to 3×10−5 M GSNO and 3×10−5 M GSNO plus oxyhaemoglobin (5×10−6 M) in random order. As before a low concentration of isoprenaline was used to enhance the response to the NO donor. GSNO produced a large increase relative to isoprenaline alone (133%) and oxyhaemoglobin abolished this effect (Fig. 2). The inhibition was reversed by removal of the oxyhaemoglogin by washing for 20 min.

Fig. 2

Effect of oxyhaemoglobin (5×10−6 M) on the response to GSNO (3×10−5 M). Contraction amplitude is displayed as a percentage of resting cell length. Each bar represents a mean±SEM (n = 7). In comparison to isoprenaline (3×10−10 M) alone, GSNO resulted in a 132±29% increase in contraction (**, P<0.005). In the presence of oxyhaemoglobin there was no response to GSNO.

3.5 Role of soluble guanylyl cyclase

Basal levels of cGMP were 1.15±0.08 pmol/200 000 viable myocytes. There was a modest rise with GSNO 10−5 M (26.6±5.3%, n = 3, P<0.05). The increase in cGMP was abolished by preincubation of myocytes with the sGC inhibitor ODQ 10−5 M (+26.6±5.3% vs. −12.6±4.9, n = 3, P<0.05). However, the functional response to GSNO, in the presence of low dose isoprenaline, was not altered by ODQ 10−5 M (Fig. 3a), nor was the response to DEA/NO in the absence of isoprenaline (amplitude increased by 108±38%, n = 4, P<0.05). The rise in cGMP was more marked with DEA/NO 10−5 M (257±98%, n = 5, P<0.05) whilst the combination of DEA/NO (10−5 M) and the sGC activator YC-1 (10−4 M) resulted in a 52±9 fold rise in cGMP levels (n = 4) showing that the cells could mount a cGMP response.

Fig. 3

(a) Concentration–effect relationship of 8-Br-cGMP on isolated guinea-pig myocytes stimulated with isoprenaline (3×10−10 M). Contractile response is displayed as a percentage of resting length. Each bar represents the mean±SEM (n = 7). There is no significant change of contraction amplitude with any dose of 8-Br-cGMP. (b) Effect of GSNO (10−5 or 3×10−5 M) on contraction in the presence or absence of the sGC inhibitor ODQ (10−5 M) (n = 6 paired experiments). GSNO showed a significant increase in contraction amplitude compared to basal contraction with isoprenaline (3×10−10 M). The response to GSNO was unaffected by ODQ.

The cGMP analogue 8-Br-cGMP (3×10−6–10−4 M) had no significant effect on cell contraction amplitude (Fig. 3b). The peak effect of 8-Br-cGMP was to depress contractility by 21.3±4.5% which just failed to reach statistical significance (P = 0.066, n = 7).

3.6 Thapsigargin

Contracting myocytes were exposed to thapsigargin (3×10−6 M) which prevents reuptake of Ca2+ into the sarcoplasmic reticulum (SR), while continued stimulation of contraction served to deplete the SR of any residual Ca2+ stores. The relative increase of contraction amplitude with 10−5 M GSNO, was unaffected (81.4±14.6%, n = 5 with thapsigargin vs. 70.9±8.3% in controls).

3.7 Role of cyclic AMP

SNAP (10−5 and 10−4 M) did not produce a statistically significant change in cAMP levels (Fig. 4a) whilst isoprenaline (10−6 M) produced more than a doubling (213±19%, P<0.001, n = 8). The most potent NO donor DEA/NO, capable of eliciting a marked inotropic effect when used alone caused a small non significant decrease in cAMP levels (6.6±11.3% of basal, n = 6).

Fig. 4

(a) Effect of SNAP (10−5 M, n = 6 and 10−4 M, n = 3), DEA/NO (n = 6) and isoprenaline (n = 8) on levels of myocyte cAMP. Levels are expressed as a percentage of basal. No significant rise was seen with an NO donor even in high concentrations; ***, P<0.001. (b) Effect of Rp-cAMPS (2×10−4 M) on the response to DEA/NO. Contractile response is displayed as a percentage of resting cell length. The inotropic response to DEA/NO (3×10−5 M) remained significant in the presence of Rp-cAMPS; **, P<0.005.

3.8 Inhibition of cAMP dependent protein kinase

Guinea-pig myocytes were treated with DEA/NO (3×10−5 M) in the absence of β-adrenergic stimulation. Contractile amplitude increased by 112±19% (n = 6 and P<0.005) compared to baseline (Fig. 4b). Cells were recirculated for 40 min with KH solution containing Rp-cAMPS (2×10−4 M). Rechallenge with DEA/NO in the presence of Rp-cAMPS produced a 119±34% increase of contraction amplitude, similar to the response seen in the absence of Rp-cAMPS.

3.9 Effect of variation in the rate of NO release

In order to explore the divergent effects of different NO donors on contractile function we examined the possible role of the rate of NO generation. Release of NO from the two NONOate donors DEA/NO and detanonoate is highly predictable and governed exclusively by first order kinetics. Total NO release is quantitatively similar but occurs at very different rates reflected in the half lives; DEA/NO has a half-life of 2 min [25] and detanonoate has a half-life of 20 h [26]. At 10−5 M DEA/NO produced a significant increase in contraction (168±59%, P<0.005, n = 5). In contrast detanonoate produced a 25.7±17.3% increase at 10−5 M (n = 8) and an increase of 66.4±38.2% at 3×10−5 M (n = 6). Five cells pretreated with low dose isoprenaline were serially exposed first to detanonoate and then DEA/NO over a concentration range of 10−6–10−4 M. The maximum change of contraction with detanonoate failed to reach significance (average 59.6±25.2%, P = NS) whilst with DEA/NO the maximum increase was 209±67% (P<0.05). Responses to DEA/NO occurred at concentrations as low as 10−6 M. It thus appeared that the faster rate of release of NO from DEA/NO is associated with a greater positive inotropic effect.

3.10 Measurement of spontaneous NO release

The rates of NO generation from the donors SNP, DEA/NO, GSNO and detanonoate (10−5–3×10−4 M) were measured directly in an acellular system (Table 1). DEA/NO, as expected, released NO rapidly. SNP, detanonoate and GSNO produced a lower but comparable release of NO at a concentration of 10−5 M, despite having different functional effects on myocyte contraction. However GSNO may be subject to metabolic breakdown [27] and therefore the rate of release was examined in a cellular system using an NO electrode (Fig. 5). In this system the NO signal was totally quenched by the scavenger Carboxy-PTIO. Liberation of NO from GSNO (10−3 M) was greatly enhanced in the presence of guinea-pig myocytes (5×105/ml). Peak NO concentration increased from 1.0±0.09 to 6.1±0.41×10−6 M (n = 4, from three different myocyte preparations, P<0.0001). The effect of cells was not due to an alteration in the pH of the medium since this was unchanged at 7.3. Pretreatment with ethylmaleimide 10−3 M (to reduce thiols) abolished the effect of myocytes to enhance NO release from GSNO. Primed medium in which cells had been suspended for more than 2 h also enhanced NO release from GSNO and this effect was abolished by adding ethylmaleimide (10−3 M) to the primed medium. Another thiol NO donor SNAP showed a similar enhancement of NO release at 10−4 M in the presence of myocytes. No enhancement of NO release from SNP (10−5–10−3 M) was produced by myocytes or primed medium (data not shown).

Fig. 5

Representative traces from an NO electrode showing the detected concentration produced by GSNO (10−3 M) alone (−cells), with 5×105 guinea-pig myocytes (+cells), with culture supernatant and with cells incubated with ethylmaleimide (10−3 M) for 10 min.

View this table:
Table 1

Spontaneous release of NO measured by dual beam spectrophotometry with results expressed as nmol/min spontaneous NO releasea

NO released (nmol/min)
10−5 M3×10−5 M10−4 M3×10−4 M
NO donor    
  • a Results are the mean of at least six observations±SEM.

  • ND ND, not determined.

3.11 Effects of GSNO and DEA/NO on human myocyte contraction

Data was collected from 14 human cells (from 13 preparations), derived from both failing and nonfailing hearts (Fig. 6). Of the eight cells exposed to GSNO, on a background of isoprenaline, two failed to respond whilst six cells showed a positive response. The overall mean increase of contraction amplitude was from 3.61±0.48 to 5.34±0.46 (% shortening), a significant increase of 48% (P<0.05). No effect of GSNO was seen on baseline contraction without isoprenaline.

Fig. 6

Effect of GSNO (3×10−5 M, n = 8) and DEA/NO (10−5 M, n = 6) on isolated human myocytes. Cell contraction amplitude is represented as a percentage of resting cell length. The mean response to GSNO was a relative increase in contraction of 48% compared with isoprenaline alone (P<0.05). The response of guinea-pig myocytes to GSNO (10−5 M), in the presence of isoprenaline (3×10−10 M) is shown for comparison. The source of the 13 human cell preparations included ischaemic cardiomyopathy (5), congenital heart disease (2), dilated cardiomyopathy (2), rejected donor organ (1) and three preparations from biopsies taken from patients undergoing routine mitral valve replacement.

In contrast DEA/NO (10−5 M) caused a marked positive effect even in the absence of isoprenaline. The small basal contraction amplitude was significantly increased (1.38±0.35 to 4.7±0.65%, n = 6, P<0.01) whilst an inotropic response was observed at concentrations as low as 10−6 M.

4 Discussion

Negative inotropic effects of endogenous and exogenous NO are well documented and it has been widely assumed that this is the major effect of NO on the heart [4,5]. However, results of this study clearly demonstrate a positive inotropic effect of certain structurally diverse NO donors. In guinea-pig myocytes this action was seen with NO-thiols (GSNO, SNAP) and NONOates (papanonoate and DEA/NO) and was reversibly blocked by oxyhaemoglobin. The effect was independent of either cGMP or calcium stored in the sarcoplasmic reticulum. Although the positive inotropic effect was enhanced by the presence of a low concentration of isoprenaline, it was not accompanied by an increase in cAMP or abolished by the cAMP dependent protein kinase inhibitor Rp-cAMPS. Human myocytes produced a similar pattern of positive inotropy to GSNO, in the presence of isoprenaline, and DEA/NO alone. The results further enforce the view that NO may cause a positive inotropic effect under certain conditions and that this could occur at therapeutically relevant concentrations of NO donors, with an important determinant being the rate of NO release.

The modest negative inotropic effect of conventional NO donors e.g. SNP has been documented in a range of mammalian species and experimental systems including isolated rat and guinea-pig myocytes. However these effects have been relatively small and not seen in all studies [17–19,28,29]. The present study explored responses of myocytes to a range of NO donors. SNP and GTN had small but nonsignificant negative inotropic effects. In contrast the thiol NO donors GSNO and SNAP, along with the NONOate donors papanonoate and DEA/NO, produced a significant increase in contraction amplitude. This inotropic response was reproducible and was seen in cells that did not respond to GTN. The effect of GSNO was not mimicked by glutathione and the positive inotropic effect appeared to be mediated by free NO since it was reversibly inhibited by oxyhaemoglobin.

We explored possible reasons why some NO donors produced a positive inotropic effect and others did not. The effect did not seem to be restricted to particular classes of NO donors that preferentially release NO+ or other NO species (e.g. NO) since it was seen with NO thiols (potential NO+ donors) and non-thiol containing donors that released NO or NO. Furthermore, the observation that the effect was abolished by oxyhaemoglobin suggests that NO is the likely mediator of the effect seen [30]. To test the hypothesis that the rate of release was important we used two compounds that contain equimolar amounts of NO and release the NO spontaneously but at vastly different rates. The positive inotropic effects of the fast releaser (DEA/NO) were much greater and more reproducible than the effects of the slow releaser (detanonoate). These data would be consistent with the rate of NO release being an important determinant of the response seen. However, when we assessed the rate of decomposition of SNP (no inotropic effect) and GSNO (positive inotrope) in an acellular system they appeared comparable. In contrast, in cellular systems, release of NO from GSNO (or SNAP) was enhanced. This observation suggests that the breakdown of GSNO is enhanced by myocytes and the consequent fast rate of NO release may account for the positive inotropic effect. However, whilst active myocytes were necessary to trigger the fast release rate of NO from GSNO, this effect was also seen if conditioned medium was added to GSNO. The stimulatory effect of conditioned medium was abolished by the presence of ethylmaleimide, a thiol binder. These results would be consistent with our previous observations that NO release from GSNO is stimulated by cysteine [31] and that the species released is NO rather than a secondary nitrosothiol. Thus the overall effect of nitrosothiols on myocyte function may be dependent on the thiol content and redox state of the cell and this may account for previous conflicting data with nitrosothiols. Indeed the conflicting reports of effects of different NO donors on myocyte function in different experimental systems might be explained by varying rates of release depending on the cell species studied and the experimental conditions.

The mechanism by which NO produces a positive inotropic effect is not clear. Although previous reports have suggested that the effect might be partially dependent on cGMP [19,32,33], in the present studies the sGC inhibitor ODQ was without effect. It has been suggested that ODQ may be relatively ineffective in rat cardiac myocytes [34], however we found, as documented by others [28], that 10−5 M ODQ blocked cGMP accumulation in guinea-pig myocytes. The basal level of cGMP in the present study was comparable to that reported previously [12] and the rise with exposure to GSNO was small though significant. A similar rise of 31±6% has been reported following exposure of rat myocytes to the thiol donor SNAP (10−4 M) [28]. The larger response to DEA/NO was very similar to the 250% rise in a previous report [35] whilst the combination of DEA/NO and YC-1 produced a marked elevation in cGMP, which has been previously observed in human platelets [36]. Despite the effect of NO donors on cGMP levels the positive inotropic effect was not mimicked by the stable cGMP analogue 8-Br-cGMP again suggesting the effect is independent of cGMP. However, it remains possible that variation in the activity of the sGC system may influence the observed response to NO donors. In rat myocytes inhibition of protein kinase G by KT5823 changed their response to DEA/NO from negative to a marked positive inotropy [28]. Thus low activity in the sGC system or reduced response of sGC to NO may allow myocytes to manifest the positive inotropy induced by activation of an alternative NO target.

The effect of NO donors also seems to be independent of cAMP. Small increases in cAMP, in response to NO donors, have been reported previously [28] in the present study neither SNAP nor the potent positive inotrope DEA/NO increased cAMP. However, even small changes in cAMP may initiate positive inotropic responses. Therefore the effect of the protein kinase A inhibitor Rp-cAMPS was studied. In contrast to Vila-Petroff et al. [28] we found no inhibitory effect at concentrations of 2×10−4 M Rp-cAMPS which we have previously shown will cause complete and reversible inhibition of the effect of maximally stimulating isoprenaline concentrations on guinea-pig myocytes [37]. Although the positive inotropic effect of NO donors was not mediated by cAMP, there was a synergism between β-adrenoceptor stimulation and GSNO response. Thus the inhibitory effects of Rp-cAMPS in the previous study may be due to interference with a tonic effect of basal cAMP production rather than a true block of NO mechanisms. The interaction of NO with adrenoceptor stimulation may be of relevance to the findings in vivo, since the raised catecholamines observed in heart failure could alter the effects of exogenous (or endogenous) NO.

An alternative target for nitrogen oxides is ion channels. Both L-type Ca2+ channels and the calcium release complex (CRC) contain sulphydryl groups which may be subject to nitrosylation or oxidation to modulate channel function and hence influence electromechanical coupling. Very high concentrations of NO-thiols cause S-nitrosylation of thiol residues of the calcium release channel (ryanodine receptor) of the sarcoplasmic reticulum and this increases the probability of a channel opening [38]. However, in our study the sarcoplasmic Ca2+ ATPase inhibitor, thapsigargin, did not block the response to GSNO and this argues against a central role for the sarcoplasmic reticulum in the response seen. The observation that extracellular haemoglobin blocked the response might suggest a cell surface target for the NO. Both SNAP (10−7 M) and spermine/NO have been shown to increase ICa in human [39] and cat atrial myocytes [40] respectively, though these effects were mediated by cGMP. In guinea-pig myocytes the combination of low dose isoprenaline and SIN-1 was capable of increasing the calcium current [41] and entry of calcium from the extracellular pool may be the mechanism underlying the observed effect. Further studies will be required to test the hypothesis directly.

The results of this study show that NO released at high rates from certain NO donors induces a positive inotropic effect in both guinea-pig and in human ventricular myocytes. Recent studies have shown a similar positive effect occurs in isolated cells [17], whole hearts [29,42], and in vivo [20,43]. There is growing evidence that the effect of NO on cardiac function is not, as initially reported, simply as a mild negative inotrope but that a range of positive and negative effects can be seen depending on the type of NO donor, the kinetics of NO release and possibly the activity or responsiveness of sGC. It would be interesting to determine to what extent these positive effects are seen in humans in vivo and whether certain NO donors have therapeutic potential for the treatment of conditions such as acute cardiogenic shock. It also remains to be determined whether endogenous thiols (GSNO, nitrosoalbumin and nitrosohaemoglobin) play a regulatory role in cardiac function in vivo.


We would like to thank Professor M. Yacoub and the transplant team at Harefield Hospital for the provision of human transplant tissue and Mr. P. O'Gara for his assistance with isolated myocyte preparation. D.S. is supported by a British Heart Foundation clinical Ph.D. studentship (FS/97060).


  1. [1]
  2. [2]
  3. [3]
  4. [4]
  5. [5]
  6. [6]
  7. [7]
  8. [8]
  9. [9]
  10. [10]
  11. [11]
  12. [12]
  13. [13]
  14. [14]
  15. [15]
  16. [16]
  17. [17]
  18. [18]
  19. [19]
  20. [20]
  21. [21]
  22. [22]
  23. [23]
  24. [24]
  25. [25]
  26. [26]
  27. [27]
  28. [28]
  29. [29]
  30. [30]
  31. [31]
  32. [32]
  33. [33]
  34. [34]
  35. [35]
  36. [36]
  37. [37]
  38. [38]
  39. [39]
  40. [40]
  41. [41]
  42. [42]
  43. [43]
View Abstract