The inotropic effects of ammonia on isolated perfused rat hearts and the mechanisms involved

Ammonia (NH₃) is a naturally occurring chemical in the atmosphere, as well as an essential man-made chemical. At room temperature, NH₃ is a colorless, pungent-smelling gas and is lighter than air. NH₃ is a major product of amino acid catabolism and profoundly affects the function of neurons and the vasculature (Li et al., 2009). With the exception of developing conceptuses and growing animals (Wu et al., 2008; Wu et al., 2009), amino acids are not stored in tissues and must be degraded to yield NH₃ in a cell-specific manner. NH₃ is also produced in the human gut by bacteria. In addition, degradation of purines and pyrimidines generates NH₃, particularly in exercising skeletal muscle (Suryawan et al., 2009).

NH₃ occurs in micromolar concentrations in human blood. It is important to note that individual blood NH₃ levels increase considerably as a result of dietary fluctuations or exercise (Dimski, 1994; Greenhaff et al., 1991). Pathophysiologically, high concentrations (≥3 mol l^–1) of NH₃ dilate human cerebrovascular smooth muscle (Andersson et al., 1981). Hyperammonemia is an important factor in the development of hepatic encephalopathy and has been suggested to trigger expression of inducible nitric oxide synthase (iNOS) in astroglial cells (Butterworth, 2002; Suárez et al., 2002) and to have dose-dependent effects on monoamine synthesis. NH₃ reduction is the target for therapy of hepatic encephalopathy (Olde Damink et al., 2002; Rudman et al., 1973; Ytrebø et al., 2006). High NH₃ levels can be lethal to the central nervous system (Wu and Morris, 1998). High NH₃ levels also induce changes in N-methyl d-aspartate (NMDA) and γ-amino butyric acid (GABA) receptors and causes downregulation in astroglial glutamate transporter molecules (Fan and Szerhb, 1993; Bender and Norenberg, 2000). In addition, like nitric oxide (NO) and hydrogen sulfide (H₂S), NH₃ may be a contributing factor in Alzheimer's disease (Seiler, 2002). A correlation between arterial NH₃ concentration and brain glutamine content in humans has been described. Moreover, brain content of glutamine is correlated with intracranial pressure. Emerging evidence shows that NH₃ itself is not harmful to the brain, but its conversion to glutamine by glutamine synthetase inhibits endothelial NOS and, therefore, blood flow and oxygen supply to the brain (Lee et al., 1996; Kawaguchi et al., 2005). In contrast, glutamine is required for production of iNOS protein by macrophages, thereby stimulating NO production under conditions of immunological activation to kill pathogenic micro-organisms (Li et al., 2007). So glutamine produced by NH₃ has different effects on NO production in different organs.

NO, H₂S, carbon monoxide (CO) and sulfur dioxide (SO₂) are endogenous gas transmitters, which affect a variety of vital functions including vasorelaxation and suppression of cell proliferation (O'Sullivan et al., 2006; Ali et al., 2006; Li and Meng, 2009; Li and Meng, 2010; Li and Moore, 2007; Zhang and Meng, 2009; Zhang and Meng, 2010). In contrast to these toxic gaseous molecules, we hypothesize that NH₃ may have a physiological role in the regulation of cardiovascular function. Currently, however, the mechanisms underlying the inotropic effects of NH₃ on isolated perfused hearts in rats are not understood. Therefore, the purpose of this study was to investigate the inotropic effects of NH₃ and its mechanisms in isolated hearts in rats.

Glibenclamide, 4-aminopyridine (4-AP), nifedipine, iberiotoxin, N^G-nitro-l-arginine methyl ester (l-NAME), 4H-8-bromo-1,2,4-oxadiazolo(3,4-d)benz(b)(1,4)-oxazin-1-one (NS-2028), staurosporine, indomethacin, propranolol and apamin were purchased from Sigma (St Louis, MO, USA). A radioimmunoassay kit for measuring cGMP was obtained from Shanghai University of Traditional Chinese Medicine (Shanghai, China). Assay kits for lactate dehydrogenase (LDH), creatine kinase (CK), NO, NOS and adenosine triphosphatase (ATPase) were purchased from Nan Jing Jian Cheng Company of Biological Technology (Nanjing, China). Other chemical reagents were of analytical grade.

Animal care and experimental protocols complied with the Animal Management Rule of the Ministry of Health, People's Republic of China (Documentation 55, 2001) and the Animal Care Committee of Shanxi University. Male Wistar rats [Rattus norvegicus (Berkenhout 1769)] weighing 220–250 g were obtained from Heibei Medical University (Shijiazhuang, China).

Rats were injected intravenously with anticoagulant heparin under the terminal anesthesia (pentobarbital sodium 30 mg kg^–1, i.p.). The hearts were quickly excised and perfused on a Langendorff apparatus with a modified Krebs–Henseleit (K–H) buffer (mmol l^–1: 118.0 NaCl, 4.7 KCl, 25.0 NaHCO₃, 1.2 KH₂PO₄, 1.2 MgCl₂, 1.25 CaCl₂ and 11.0 glucose, pH 7.3) under a 100 cm H₂O pressure at 37°C, gassed with 95% O₂–5% CO₂. A saline-filled latex balloon was inserted into the left ventricle through the left atrium and connected to a pressure transducer for measurement of left ventricular pressure. Heart rate, coronary flow, left ventricular developed pressure (LVDP=left ventricular systolic pressure–left ventricular diastolic pressure) and maximum left ventricular pressure development (±LV dP/dt_max) were measured by a MedLab Biological Signal Collection System (Medease Science and Technology, Nanjing, China) with the left ventricular diastolic pressure pro-stabilized at 10 mmHg (Suzuki et al., 2000). The whole system was water jacketed and maintained at 37°C. The hearts were perfused for 20–30 min to establish equilibrium hemodynamics. The equilibrium phase was terminated when LVDP, heart rate and coronary flow were maintained at the same level for three continuous periods of measurement, 5 min apart. The baseline measurements were recorded at the end of this time. Hearts not meeting these criteria were not used in the study.

To study the positive inotropic effect of NH₃ on the isolated perfused rat hearts, NH (0–4 mmol l^–1₃) was added to the perfused fluid. The parameters of cardiac function under various conditions were measured. In order to study whether the impact of NH₃ on heart function was related to the change in pH value of the perfusate or to NH ⁺₄, we also studied the effects of NaOH and NH₄Cl (0–4 mmol l^–1) on heart function.

To investigate the roles of Ca²⁺ and K⁺ channels and the various cellular signaling pathways in the effect of NH₃ on the isolated rat hearts, various specific inhibitors (nifedipine 30 nmol l^–1, glibenclamide 10 mmol l^–1, 4-AP 2 mmol l^–1, iberiotoxin 100 nmol l^–1, l-NAME 100 mmol l^–1, NS-2028 10 mmol l^–1, propranolol 10 mmol l^–1, indomethacin 10 mmol l^–1, staurosporine 30 nmol l^–1, apamin 50 nmol l^–1 and saline control) were applied in this study. Hearts were allowed to stabilize for 20–30 min, and then inhibitors were administered for a 10 min pre-dosing period before the addition of NH₃. Inhibitors and NH₃ (0.5 or 1.5 mmol l^–1) were then administered together for a 10 min period. Heart rate, coronary flow, LVDP and ±LV dP/dt_max were measured throughout the experiments. The data for this set were calculated as the percentage of pre-NH₃ values.

Perfusate was collected during the NH₃ perfusion period. The activities of LDH and CK in the perfusate were assayed using a kit according to the manufacturer's instructions.

Rat hearts were allowed to stabilize for 20–30 min, and then NH₃ (0.5 or 1.5 mmol l^–1) was added to the perfused fluid for 10 min. The hearts were quickly frozen in liquid nitrogen, and subsequently used for measuring the activity of ATPase and NOS and the level of NO and cGMP.

Total NOS (tNOS) and inducible NOS (iNOS) activity was measured by using a NOS assay kit according to the manufacturer's protocol. The resulting absorbance was determined at 530 nm with a spectrophotometer.

Ca²⁺,Mg²⁺-ATPase and Na⁺,K⁺-ATPase activity was estimated by quantifying the release of inorganic phosphorus from adenosine triphosphate (ATP) according to the kit instructions. One activity unit of ATPase is expressed as 1 μmol phosphate mg^–1 protein h^–1.

To measure the content of cGMP, the perfused hearts were homogenized in a glass homogenizer. The homogenates were centrifuged at 1000 g for 10 min at 4°C and the supernatants were extracted with two volumes of water-saturated ethanol and 75% ethanol. The supernatant was taken for cGMP measurement using a radioimmunoassay kit. The radioactivity present in the bound fraction was measured in terms of counts per minute (c.p.m.) by a Gamma Counter (FT-646A, Beijing, China). The concentrations of cGMP were calculated by interpolation of %B/B₀ (where B represents the c.p.m. value of the sample tube and B₀ is the c.p.m. value of the blank tube) from a standard curve and are expressed as pmol g^–1.

NO levels were measured using a NO assay kit according to the manufacturer's protocol. The resulting absorbance was determined at 550 nm with a spectrophotometer.

Protein content of the myocardial homogenate was evaluated by the Bradford method (Bradford, 1976), with bovine serum albumin used as a standard.

All values are expressed as means ± s.d. Student's t-test for unpaired samples was used to compare the mean values between the control and test groups. Multiple comparisons were made with one-way ANOVA followed by post hoc analysis (Tukey's test). Statistical significance was set at P<0.05.

Fig. 1 and Table 1 show the changes in ±LV dP/dt_max, LVDP, heart rate and coronary flow of isolated hearts exposed to the different concentrations of NH₃, NaOH and NH₄Cl. These three substances all elicited positive inotropic effects in a concentration-dependent manner, with the maximum response shown at the highest concentration tested (4 mmol l^–1). NH₃ produced a higher positive inotropic effect than NaOH and NH₄Cl at the same concentration. From 0.5 to 4 mmol l^–1, the increases in LVDP induced by NH₃ were significantly greater than those induced by NaOH and NH₄Cl (Fig. 1). In addition, the heart rate and coronary flow were significantly increased by NH₃ at concentrations from 0.5 to 4 mmol l^–1, but they only were significantly increased by NaOH at the highest concentration tested (4 mmol l^–1). A concentration of 0.1 mmol l^–1 NH₃ was infused for 10 min resulting in a statistically significant increase in +LV dP/dt_max and –LV dP/dt_max, by 10.14–89.97% and 12.82–95.68%, respectively (Table 1). In contrast, no significant changes in the ±LV dP/dt_max were observed with 0.1 mmol l^–1 NaOH or NH₄Cl. Table 2 presents changes of pH value of K–H buffer caused by NH₃ and NaOH at different concentrations at 37°C. At a concentration of 0.5 mmol l^–1, NH₃ did not change the pH value of K–H buffer. At higher concentrations (2 and 4 mmol l^–1), the increases in the pH value of the K–H buffer induced by NH₃ were significantly lower than those induced by NaOH at the same concentration.

The role of the ATP-sensitive K⁺ (K_ATP) channel in the NH₃-mediated inotropic effect was examined by administering glibenclamide (10 mmol l^–1) before and during NH₃ infusion. Glibenclamide alone significantly decreased coronary flow by around 15% over the baseline value but did not affect LVDP and heart rate. In the presence of glibenclamide, the increases in LVDP induced by NH₃ (0.5 and 1.5 mmol l^–1) were partially inhibited (Fig. 2).

The role of NO synthase in the NH₃-mediated inotropic action was examined by administering l-NAME (100 mmol l^–1) before and during NH₃ infusion. l-NAME alone significantly decreased LVDP by around 15% over the baseline value but did not have a significant effect on the coronary flow and heart rate. In the presence of l-NAME, the increases in LVDP induced by NH₃ (0.5 and 1.5 mmol l^–1) were partially inhibited (Fig. 3).

NS-2028 has been shown to specifically inhibit soluble guanylate cyclase and reduce the production of cGMP (Olesen et al., 1998). In this study, NS-2028 (10 mmol l^–1) was used to treat isolated hearts 10 min before the application of NH₃. NS-2028 alone significantly decreased LVDP and coronary flow by around 15% over the baseline value but did not have any effect on the heart rate. The increases in LVDP induced by NH₃ (0.5 and 1.5 mmol l^–1) were partially inhibited by NS-2028 (Fig. 4).

To determine the roles of the BK_Ca (big Ca²⁺-activated K⁺) channel, SK_Ca (small Ca²⁺-activated K⁺) channel, K_v (voltage-dependent K⁺) channel, L-type calcium channel, protein kinase C (PKC), cycloxygenase and β-adrenoceptor in the positive inotropic effect of NH₃, the hearts received iberiotoxin, apamin, 4-AP, nifedipine, staurosporine, indomethacin and propranolol, respectively, before and during NH₃ infusion. However, the positive inotropic effect of NH₃ in the hearts was not affected by these drugs (data not shown).

The role of the K_ATP channel and SK_Ca channel in the NaOH-mediated inotropic action was examined by administering either glibenclamide (10 mmol l^–1) or apamin (50 nmol l^–1) before and during the NaOH infusion. Glibenclamide alone significantly decreased coronary flow by around 15% over the baseline value but did not have any effect on LVDP and heart rate. Apamin alone did not have any effect on LVDP, coronary flow and heart rate. In the presence of glibenclamide or apamin, the increases of LVDP induced by NaOH at 1 mmol l^–1 were partially inhibited (Fig. 5).

However, the positive inotropic effect of NaOH on the hearts was not affected by other drugs used in our paper (data not shown).

From Fig. 6 we can see that the activity of LDH and of CK in the coronary perfusate was significantly higher in the 1.5 mmol l^–1 NH₃ group than in the control group.

Fig. 7 presents the activity of NOS and ATPase in rat hearts perfused with 0.5 and 1.5 mmol l^–1 NH₃. Both tNOS activity and iNOS activity in the hearts were significantly increased by 0.5 and 1.5 mmol l^–1 NH₃. In addition, at a concentration of 1.5 mmol l^–1, NH₃ caused significant decreases in Na⁺,K⁺-ATPase and Ca²⁺,Mg²⁺-ATPase activity in the hearts.

NO and cGMP levels were assessed in rat hearts perfused with NH₃ (Fig. 8). Levels of both NO and cGMP in the hearts were significantly increased by 0.5 and 1.5 mmol l^–1 NH₃.

In this study, we utilized the isolated perfused heart model to elucidate the effect of NH₃ on myocardial contractility and hemodynamic parameters. The results show that NH₃ elicits a positive inotropic effect in a concentration-dependent manner. This effect cannot be explained by changes in pH or NH ⁺₄ alone as NH₃ produced a higher positive effect than NaOH and NH₄Cl at the same concentration. In particular, at a concentration of 0.5 mmol l^–1, NH₃ did not change the pH value of K–H buffer at 37°C (Table 2), but NH₃ elicited a significant positive inotropic effect. From 0.5 to 4 mmol l^–1, the increases in LVDP induced by NH₃ were significantly higher than those induced by NaOH and NH₄Cl (Fig. 1). In addition, heart rate and coronary flow were significantly increased by NH₃ at concentrations from 0.5 to 4 mmol l^–1, but were only significantly increased by NaOH at the highest concentration tested (4 mmol l^–1). A concentration of 0.1 mmol l^–1 NH₃ infused for 10 min resulted in a statistically significant increase in the ±LV dP/dt_max. In contrast, no significant changes in the ±LV dP/dt_max were observed with 0.1 mmol l^–1 NaOH or NH₄Cl (Table 1). These findings suggest that, at low concentrations, the impact of NH₃ on cardiac function was the result of NH₃ molecules; at high concentrations, the effect of NH₃ on the hearts might be partly correlative with the change of pH value, but mainly caused by NH₃ molecules. One possible reason is that NH₃ molecules pass membrane barriers by diffusion more easily than NH ⁺₄ and OH^– (Cooper, 1994; Kvamme, 1983). In addition, another study from our laboratory indicated that NH₃ at low concentrations could promote a concentration-dependent vasoconstriction in the isolated aorta in rats (Z.M., unpublished).

The opening of the K_ATP channel is associated with potassium efflux, polarization of the cell membrane and shortening of action potential duration (Noma, 1983; Cole et al., 1991). These effects reduce Ca²⁺ influx via L-type Ca²⁺ channels and decrease the time for Na⁺–Ca²⁺ exchange operating in the reverse mode. The resulting decrease in Ca²⁺ influx would be expected to lead to a reduction of cardiomyocyte contraction. We determined whether the K_ATP channels were involved in the positive inotropic effects of NH₃. Glibenclamide, a K_ATP channel inhibitor, partially inhibited the positive effects induced by NH₃ (0.5 and 1.5 mmol l^–1). These results suggest that the positive inotropic effects of NH₃ might be related to K_ATP channels. The mechanism behind this observation is unclear and further research is needed.

In the present study, l-NAME and NS-2028 partially attenuated the positive inotropic effects induced by the different concentrations of NH₃. In addition, NO and cGMP levels in the heart were significantly increased by different concentrations of NH₃. The endogenously produced small increase in NO and cGMP can elicit a positive inotropic effect in isolated rat hearts (Kojda et al., 1997; Massion et al., 2003). These results suggest that NO and cGMP play a significant role in the NH₃-mediated positive inotropic effect. Our results are consistent with those of a previous report by Faff et al. (Faff et al., 1996), who proposed that NH₃ might act in cell signaling through activation of cAMP-dependent signaling pathways in which mildly increased cGMP inhibits cAMP hydrolysis through inhibition of phosphodiesterase.

Furthermore, we investigated the mechanism of the positive inotropic effect induced by NaOH on rat hearts. The positive inotropic effect of NaOH at 1 mmol l^–1 was partially inhibited by glibenclamide and apamin. This result suggests that the effect was partially related to the K_ATP channel and SK_Ca channel. Although the positive inotropic effect of NH₃ was also partially related to K_ATP channels, the mechanism differed from that for NaOH. These results further indicate that the positive inotropic effect induced by NH₃ was mainly caused by NH₃ molecules and was partly correlated with the change of pH value of the heart perfusate. Similarly, Andersson and colleagues (Andersson et al., 1981) reported that the relaxation of cerebrovascular smooth muscle caused by NH₃ was independent of NH₃-induced extracellular changes in pH. Wakabayashi and colleagues (Wakabayashi et al., 1992) reported that ammonium compounds increased vascular tone by causing influx of extracellular calcium through the voltage-dependent calcium channel, and intracellular alkalinization was involved in this process.

CK is an enzyme (EC 2.7.3.2) expressed by various tissues and cell types. CK (using ATP) catalyzes the conversion of creatine to phosphocreatine and adenosine diphosphate (ADP). Thus CK is an important enzyme in many tissues (Wallimann and Hemmer, 1994). Increased CK levels may signal muscular dystrophy, polymyositis, dermatomyositis and sometimes myositis, as well as heart attacks. An increase of extracellular LDH enzyme activity reflects an increase in the number of membrane-damaged cells (Lobner, 2000). In addition, the plasma membrane Ca²⁺,Mg²⁺-ATPase and Na⁺,K⁺-ATPase play important roles in maintaining the normal active transport for positive ions such as Na⁺, K⁺ and Ca²⁺. Inhibition of these enzymes may lead to the elevation of plasma membrane permeability for positive ions, possibly resulting in a drop in intracellular K⁺ content and an abnormal rise of cytosolic Ca²⁺ concentration (Jurma et al., 1997). In the present study, at a concentration of 1.5 mmol l^–1, NH₃ significantly increased CK and LDH activity in the coronary perfusate and significantly decreased the activity of Na⁺,K⁺-ATPase and Ca²⁺,Mg²⁺-ATPase in the heart. These results suggested that NH₃ at a concentration of 1.5 mmol l^–1 has a damaging effect on the isolated hearts.

In conclusion, we have shown that NH₃ produces a positive inotropic effect in a concentration-dependent manner. At low concentrations, the impact of NH₃ on cardiac function was via NH₃ molecules; at high concentrations, the effect of NH₃ on the hearts might be partly correlated with the change of pH value, but mainly caused by NH₃ molecules. The mechanism of the positive inotropic effect induced by NH₃ on the hearts might be related to the K_ATP channel and activation of the NO–cGMP pathway. These results suggest that NH₃ at physiological or low concentrations might play a modulatory role in heart function, but at high concentrations has a damaging effect on isolated rat hearts.

 
• ±LV dP/dt_max

  maximum left ventricular pressure development

 
• 4-AP

  4-aminopyridine

 
• ATPase

  adenosine triphosphatase

 
• BK_Ca

  big Ca²⁺-activated K⁺ channel

 
• CK

  creatine kinase

 
• GABA

  γ-amino butyric acid

 
• H₂S

  hydrogen sulfide

 
• K_ATP

  ATP-sensitive K⁺ channel

 
• K_v

  voltage-dependent K⁺ channel

 
• LDH

  lactate dehydrogenase

 
• l-NAME

  N^G-nitro-l-arginine methyl ester

 
• LVDP

  left ventricular developed pressure

 
• NH₃

  ammonia

 
• NMDA

  N-methyl d-aspartate

 
• NOS

  nitric oxide synthase

 
• NS-2028

  4H-8-bromo-1,2,4-oxadiazolo(3,4-d)benz(b)(1,4)-oxazin-1-one

 
• SK_Ca

  small Ca²⁺-activated K⁺ channel