Nitric Oxide Production and Endothelium-Dependent Vasorelaxation Ameliorated by N1-Methylnicotinamide in Human Blood Vessels
N1-methylnicotinamide (MNA+) has until recently been thought to be a biologically inactive product of nicotinamide metabolism in the pyridine nucleotides pathway. However, the latest observations imply that MNA+ may exert antithrombotic and anti-inflammatory effects through direct action on the endothelium. We examined both in vivo and in vitro whether the compound might induce vasorelaxation in human blood vessels through the improvement of nitric oxide (NO) bioavailability and a reduction of oxidative stress mediated by endothelial NO synthase (eNOS) function. MNA+ treatment (100 mg/m2 orally) in healthy normocholesterolemic and hypercholesterolemic subjects increased the l-arginine (l-NMMA)-inhibitable flow-mediated dilation (FMD) of brachial artery responses that also positively correlated with MNA+ plasma concentrations (r=0.73 for normocholesterolemics and r=0.78 for hypercholesterolemics; P<0.0001). MNA+ increased FMD at the same concentration range at which it enhanced NO release from cultured human endothelial cells after stimulation with either the receptor-dependent (acetylcholine) or the receptor-independent endothelial NO synthase agonists (calcium ionophore A23187). MNA+ restored the endothelial NO synthase agonist-stimulated NO release after the exposure of the cells to oxidized low-density lipoprotein. This effect was also associated with the normalization of the [NO]/[superoxide] balance in the endothelial cells. Taken together, the increased NO bioavailability in the endothelium contributes to the vasorelaxating properties of MNA+. Targeting eNOS with MNA+ might be therapeutically relevant for functional disorders of the endothelium, such as hypercholesterolemia and atherosclerosis.
- endothelial cells
- nitric oxide
- endothelial nitric oxide synthase
- oxidized low-density lipoprotein
- flow-mediated dilation
N1-methylnicotinamide (NMA+) is an endogenous organic cation that is biosynthesized in the liver from exogenous niacin and tryptophan via nicotinamide (NA) in a reaction catalyzed by NA N-methyltransferase.1 In contrast to NA, MNA+ has until recently been construed to be biologically inactive.2 However, the most recent studies suggest an antithrombotic activity of MNA+ related to both the inhibition of platelet aggregation and the activation of fibrinolysis, offering certain advantages over the use of NA.3–6 These observations have been provisionally attributed to the release of endothelial mediators such as prostacyclin, which boasts antiaggregatory and profibrinolytic properties. However, the actual mechanism of MNA+ effects still begs comprehensive clarification.
An improvement in the dysfunctional endothelial NO synthase (eNOS)/NO pathway is an attractive strategy in preventing and treating cardiovascular diseases.7–10 Although NO bioavailability is decreased in dysfunctional endothelium, the levels of eNOS mRNA and protein are maintained or even enhanced but associated with the increased NO synthase-dependent superoxide (O2·−) formation because of the enzymatic “uncoupling” of NO synthase; electron flow through the eNOS enzyme is then diverted to molecular oxygen rather than to l-arginine, which facilitates the production of O2·− rather than NO.7,9–11 This consequently leads to an O2·− reaction with NO, resulting in the formation of highly reactive and cytotoxic peroxynitrite and the loss of NO bioavailability.
We hypothesized that the aforementioned vascular effects of MNA+ may be mediated by the changes in eNOS-derived NO. We, therefore, investigated the effect of MNA+ on eNOS-dependent endothelial function using the ultramicrosensors for the measurement of biologically active NO with concurrent measurements of O2·− in real time in a single endothelial cell. The NO and O2·− ultramicrosensors, designed for cell cultures, allow for direct quantification of both of these radical species with high sensitivity.12,13 This approach is particularly useful for testing the compounds that potentially target eNOS.10 In clinical terms, there is a prevalent notion that endothelium-dependent vasodilation is to be regarded as a surrogate for NO bioavailability. This also prompted us to assess endothelial function as an endothelium-dependent, flow-mediated dilation (FMD) of the brachial artery in healthy normocholesterolemic and hypercholesterolemic humans treated with the pharmacological dose of MNA+.
Materials and Methods
Details are available in the online-only Data Supplement.
Study Subjects and Experimental Protocol
We examined in the double-blind, randomized, placebo-controlled study the effect of oral MNA+ administration on endothelial function in healthy normocholesterolemic (age: 32.4±9.6 years; n=16), and hypercholesterolemic subjects (age: 31.6±8.2 years; n=24) with low-density lipoprotein (LDL) cholesterol >3.4 mmol/L (Table S1, available in the online-only Data Supplement).
Study subjects were randomly assigned to receive either MNA+ (100 mg/m2 of body surface area; 2.43±0.28 mg/kg of body weight) or organoleptically identical placebo (tablets of microcrystalline cellulose as a vehicle) with a 1:1 allocation ratio. Endothelium-dependent, FMD of brachial artery in response to reactive hyperemia and endothelium-independent, nitroglycerin-induced dilation (NTG-MD) was evaluated noninvasively by the use of high-resolution ultrasound before, 2 and 4 hours after oral administration of MNA+ or placebo, during infusion (into the brachial artery) of saline (0.9% NaCl vehicle) or NG-monomethyl-l-arginine (l-NMMA), a selective inhibitor of NO synthase.14
The study was approved by the local ethics review committees of the Jagiellonian University School of Medicine and the Medical University of Gdansk.
Determination of MNA+
Blood samples were obtained from the antecubital vein before and 2 and 4 hours after oral administration of the compound. Concentrations of MNA+ in the plasma blood samples were measured by high-performance liquid chromatography with fluorescent detection.15
Cell Culture and Treatments
Human umbilical vein endothelial-derived E.A.hy926 cells after obtaining confluence (4 to 5×105 cells per 35-mm dish) were used for electrochemical measurements of NO and O2·−. Before the measurements, the cells were pretreated with different concentrations of MNA+ for 1 to 180 minutes with or without NG-nitro-l-arginine methyl ester (l-NAME), a selective eNOS inhibitor, oxidized LDL (ox-LDL), or a combination of ox-LDL and MNA+.
NO and O2·− Measurements
A 3-electrode system was used for concurrent measurements of NO and O2·−, consisting of the NO and O2·− ultrasensor working electrodes combined into a single working module, a silver/silver chloride reference electrode, and a platinum-wire counter electrode.10,12 A E.A.hy926 culture cluster was placed in a well on the stage of an inverted research microscope (Olympus, IX81) equipped with digital camera. A module of NO/O2·− ultramicrosensors was lowered near the surface (5±2 μm) of a single cell membrane with the aid of a computer-controlled micromanipulator. To stimulate NO and O2·− releases dependent on eNOS activation, the receptor-dependent acetylcholine (Ach) and the receptor-independent eNOS agonist calcium ionophore (CaI) A23187 were then injected with a microinjector, also positioned by a computer-controlled micromanipulator.
Calculation and Statistical Analysis
All of the results are reported as mean±SD. Statistical evaluation was pursued with the aid of ANOVA, followed by the Student t test. The Spearman rank correlation test was used to calculate the correlation coefficient between plasma MNA+ concentration and FMD. The value of P<0.05 was considered statistically significant.
Endothelium-Dependent FMD and MNA+ Plasma Concentration
Endothelium-dependent FMD (Figure 1A and 1B), but not endothelium-independent NTG-MD (Figure 1C and 1D), was significantly lowered in the hypercholesterolemics, as compared with the normocholesterolemic controls, before the treatment with both MNA+ (Figure 1A and 1C; 3.84±1.01% versus 6.52±1.87%, respectively; P<0.001) and placebo (Figure 1B and 1D). In either group, infusion of l-NMMA during pretreatment with MNA+ almost completely abolished the FMD response (6.52±1.87% without l-NMMA versus 0.19±0.86% with l-NMMA for normocholesterolemic and 3.84±1.01% without l-NMMA versus 0.14±0.85% with l-NMMA for hypercholesterolemic subjects; P<0.0001 for each group). In contrast, l-NMMA infusion did not affect the NTG-MD response, thus confirming that the differences in FMD between hypercholesterolemic and normocholesterolemic subjects were associated with the changes in eNOS activity and NO generation. No differences in blood pressure or heart rate were observed during the study between the MNA+- and the placebo-treated groups.
As shown in Figure 1A, FMD of the brachial artery was significantly elevated 2 and 4 hours after MNA+ administration in both studied groups (6.5±1.9% at baseline versus 10.7±3.1% at 2 hours and 10.5±1.5% at 4 hours after MNA+ in normocholesterolemics and 3.8±1.0% at baseline versus 10.7±2.1% at 2 hours and 10.6±1.8% at 4 hours after MNA+ in hypercholesterolemics; P<0.001 for each time point versus baseline in both groups). The increase in FMD after the administration of the compound was more pronounced in hypercholesterolemic subjects than in normocholesterolemic subjects (190±79% versus 68±14% at 2 hours after MNA+ and 187±59% versus 75±18% at 4 hours after MNA+ in normocholesterolemic and hypercholesterolemic subjects, respectively; P<0.01). In both groups, l-NMMA almost completely inhibited FMD, with no difference between previous and subsequent administration of MNA+. Essentially, NTG-MD did not change within 4 hours after administration of MNA+ and was not affected by l-NMMA (Figure 1C). No differences were detected between either FMD (Figure 1B) or NTG-MD (Figure 1D) before and 2 and 4 hours after placebo administration.
Plasma MNA+ concentrations increased significantly after an oral administration of MNA+, from 0.019±0.008 μmol/L at baseline to 0.122±0.075 μmol/L at 2 hours and 0.090±0.051 μmol/L at 4 hours in normocholesterolemic subjects and from 0.021±0.006 μmol/L at baseline to 0.141±0.043 μmol/L at 2 hours and 0.118±0.042 μmol/L at 4 hours in hypercholesterolemic subjects (P<0.01 for each time point versus baseline in both groups; Figure 1E). The concentration of MNA+ was affected by neither l-NMMA nor placebo (Figure 1E and 1F).
Regression analyses were carried out to further evaluate the relationship between MNA+ and endothelial function. A highly significant correlation was found between plasma MNA+ concentrations and FMD in both studied groups (r=0.73 in normocholesterolemic subjects and r=0.78 in hypercholesterolemic subjects; P<0.0001 for each group; Figure 2). No difference in association between these 2 parameters was observed in either group in the data set obtained either 2 hours or 4 hours after MNA+ administration (P<0.0001 for each time point).
Effect of MNA+ on NO Production via eNOS
We then evaluated the effect of MNA+ on endothelium-dependent NO release from human culture endothelial cells as a function of dosage and incubation time. MNA+, per se, did not stimulate NO release from endothelial cells. After Ach or CaI administration, a rapid release of NO was observed from both the control and the MNA+-treated cells. The incubation of cells for 30 minutes with increasing MNA+ concentrations resulted in a dose-dependent increase of the peak NO production after stimulation of eNOS by either CaI or Ach (Figure 3A).
The curves reflecting the production of the species reached a semiplateau at ≈10 μmol/L of MNA+ for each of the eNOS activators (539±28 nmol/L after Ach and 875±40 nmol/L after CaI). Subsequently, the peak NO concentrations did not change significantly at MNA+ concentrations ≤100 μmol/L (the highest MNA+ concentration tested). An analysis of the time-dependent effect of the constant MNA+ concentration on the stimulated peak NO release showed a rapid rise of NO production for 30 minutes after the commencement of cell exposure to the compound (Figure 3B). This effect on the agonist-stimulated NO production was maintained for ≤180 minutes from the onset of cell incubation with MNA+.
At concentrations of 0.5 to 10.0 μmol/L, MNA+ had a concentration-dependent potentiating effect on both Ach- and CaI-stimulated NO release from endothelial cells, shifted the eNOS agonists-concentration response curves to the left, and increased the maximal NO releasing response (Figure 4A for Ach and Figure 4B for CaI). To verify the potential involvement of eNOS on the effect of MNA+ on Ach- and CaI-stimulated NO release, endothelial cells were incubated with different concentrations of MNA+ in combination with 300 μmol/L of l-NAME before the administration of the eNOS agonists. l-NAME significantly decreased the release of NO throughout the concentration response to both Ach and CaI.
Effect of MNA+ on eNOS Coupling/Uncoupling
To directly assess the effect of MNA+ on NO bioavailability and eNOS functional state, the kinetics of NO release were determined with concurrent kinetics of O2·− release after the stimulation of eNOS with CaI in the single endothelial cells using tandem NO/O2·−-ultramicrosensors. The effect of MNA+ was examined in the nonox-LDL–treated cells, as well as in the ox-LDL–treated cells that were used as a model of endothelial dysfunction (Figure 5A for NO and Figure 5B for O2·−). Both concentration profiles changed with time, and the maximal concentrations changed appreciably after MNA+ incubation with either the nonox-LDL–treated or the ox-LDL–treated cells. The presence of MNA+ potentiated the increase in the rate of NO release, as well as the peaks of NO concentrations achieved after the stimulation with CaI in both cell treatment models. Furthermore, the incubation of the cells with MNA+ resulted in the kinetics of O2·− release in ox-LDL–treated cells being similar to those in the nonox-LDL–treated cells.
Because the changes in kinetics of NO and O2·− releases could be related to O2·− scavenging properties of MNA+, per se, we also examined the effect of MNA+ on O2·− released in the xanthine/xanthine oxidase system (Figure S1, available in the online-only Data Supplement). MNA+ did not reveal O2·− scavenging properties in the concentrations that affected the kinetics of NO and O2·− releases in the cells after eNOS stimulation.
The analysis of CaI-stimulated peak responses of NO and O2·− concentrations after cell incubation with MNA+ or l-NAME in the nonox-LDL–treated and the ox-LDL–treated cells is shown in Figure 6. The set of experiments with l-NAME clearly confirmed that the release of NO and O2·− observed after adding CaI was attributed to eNOS activation, because eNOS inhibition of the enzyme by l-NAME significantly blocked the release of both of the detected species. In comparison with the nontreated control cells, the cells treated with ox-LDL significantly revealed lowered CaI-stimulated peak NO concentration with an accompanying elevation of the maximal concentration of O2·−. The incubation of ox-LDL–treated cells with MNA+ restored CaI-stimulated peak NO concentration to the level observed in the nonox-LDL–treated control cells (614±38 versus 561±41 nmol/L; P value not significant). In the presence of MNA+, there were no differences in CaI-stimulated peak O2·− concentrations between ox-LDL–treated and the nontreated cells (30±2.5 versus 32±3 nmol/L; P value not significant).
Intriguingly, although MNA+ significantly increased the peak concentrations of both NO and O2·− after stimulation with CaI in the nonox-LDL–treated cells, the ratio of NO:O2·− concentration remained unchanged (26±1.4 versus 25.5±1.8 with and without MNA+, respectively; P value not significant; Figure 7, left 2 bars). In contrast to the control cells, the incubation of the ox-LDL–treated cells with MNA+ significantly increased not only the peak concentrations of both NO and O2·− after stimulation with CaI but also elevated the [NO]:[O2·−] ratio (20.5±1.4 versus 4.4±0.6 with and without MNA+, respectively; P<0.001; Figure 7, right 2 bars). We used the [NO]:[O2·−] ratio to quantify the relation between bioactive NO and cytotoxic O2·− within the endothelium that was attributed to eNOS activation. The [NO]:[O2·−] ratio was, therefore, considered an indicator of eNOS coupling/uncoupling and a marker of endothelial function/dysfunction. Thus, the presence of MNA+ prevented eNOS uncoupling in the nonox-LDL control cells (an increase of NO generation was not associated with a relative increase in release of O2·− derived from eNOS), whereas it restored eNOS coupling (an increase of NO generation was associated with a relative decrease in the release of O2·− derived from eNOS) in ox-LDL–treated cells.
Endothelial dysfunction construed as a decrease in NO bioavailability in the vessels is a common and early pathogenic mechanism by which different cardiovascular risk factors cause atherosclerotic vascular damage, predisposing patients to cardiovascular events.7–10 It is, therefore, important to seek out effective strategies aimed at preventing or treating endothelial dysfunction via increasing the bioavailability of NO derived from eNOS. Our study reports for the first time that MNA+, a primary metabolite of NA, is potent in increasing eNOS-mediated NO release from the endothelial cells. We demonstrated this effect in normal and hypercholesterolemic humans (at the early stages of vascular disease) in which the abnormality in vasodilator function was confined to the endothelium- and NO-dependent mechanisms. In contrast to NTG-MD, FMD, after a brief period of reactive hyperemia fully inhibited by l-NMMA, was significantly impaired in the hypercholesterolemic, as compared with the normocholesterolemic, subjects. The reduced portion of FMD inhibited by l-NMMA in hypercholesterolemia (compared with normal subjects) implies that the endothelial release of NO is significantly impaired in the hypercholesterolemic subjects.
By using the NO porphyrinic ultramicrosensor, it was possible to show that MNA+-enhanced eNOS agonist-stimulated NO release in human endothelial cells at potential therapeutic concentrations effectively improved the l-NMMA–inhibitable FMD response in healthy normocholesterolemic and hypercholesterolemic subjects. MNA+ was capable of normalizing a functional injury to the endothelium (before morphological lesions develop) caused by elevated LDL cholesterol levels to the extent observed in the normocholesterolemic subjects. The increase in NO release occurred at MNA+ concentrations that were similar to the plasma concentrations obtained in the subjects after single dosages of 180 to 210 mg of MNA+. Furthermore, a strong positive linear correlation was observed between the plasma concentrations of MNA+ and the extent of FMD response after treatment with the compound, confirming the in vitro observations that MNA+ directly acted on the endothelium through an increase of NO bioavailability. This is consistent with the previous studies in which MNA+ inhibited platelet aggregation in the vasculature of hypertensive animals and the effect was reversed by the eNOS inhibitor l-NAME, whereas MNA+ failed to influence the in vitro platelet aggregation.6
MNA+ has been reported to improve the endothelium- and NO-dependent vasodilation impaired in hypertriglyceridemic and diabetic animals.4 Considering the potential concentrations of MNA+ in blood that are required to reach its therapeutic effect of vasoprotection, it is worth noting that the ionic character of the MNA+ compound, which, when introduced into blood circulation, is capable of interacting with glycosaminoglycans located on the cell surface of vascular endothelium, may lead to increasing its local concentration in the vicinity of the cell membranes.2,16
In our studies, the release of bioactive NO from the endothelial cells after previous exposure to MNA+ was significantly enhanced in response to either Ach or CaI, a receptor-independent and a receptor-dependent eNOS agonist, respectively. This suggests that MNA+ improved NO bioavailability in response to the eNOS agonists by the mechanism unrelated to the muscarinic cell receptors but affected directly by the eNOS function. The release of NO after stimulation with eNOS agonists in the presence of MNA+ was inhibited by l-NAME, and the actual extent of inhibition (by 75% to 90% for each agonist) did not differ from that observed when the cells were pretreated without MNA+ (data not shown). The extent of inhibition is typical for that observed when NO generated by eNOS is being detected close to the surface of an endothelial cell by porphyrinic ultramicrosensor.10,12 This observation corroborates the fact that MNA+ improved the eNOS agonist-stimulated NO release by altering the eNOS function.
eNOS uncoupling accompanies numerous common diseases, for example, hypercholesterolemia, hypertension, and diabetes mellitus.7–10 We have demonstrated recently that eNOS uncoupling occurs not only under specific pathological conditions but also after enzyme activation in the normal endothelium.12 Rapid release of NO by most eNOS agonists is always followed by the release of O2·−, for example, CaI and Ach. Therefore, in some of our experiments the release of NO and O2·− was measured simultaneously, because O2·− generation during NO production is a major determinant of bioavailability of diffusible NO. Real-time measurements of NO and O2·− released in a single endothelial cell with tandem ultramicrosensors revealed that a short time exposure to MNA+ of normal, functional cells may lead to an increase in NO bioavailability with no changes in the NO/O2·− balance after eNOS stimulation. Furthermore, MNA+ may favorably shift the NO/O2·− balance after eNOS stimulation in the highly dysfunctional cells pretreated with ox-LDL. In the dysfunctional cells, MNA+ increased the level of bioactive NO and reduced the level of O2·−, the primary component of oxidative stress. This implies that MNA+ may be considered as a useful agent in clinical application for either the prevention of endothelial dysfunction by preserving eNOS coupling (in normal endothelial cells) or the restoration of endothelial function by the reversal of eNOS uncoupling (in the endothelial cells exposed to a risk factor).
Notably, not only did MNA+ enhance the ratio of bioactive NO versus O2·− but also the rate of NO generation, and it reduced the rate of NO fading. The favorable changes of MNA+ in the kinetics of NO release after stimulation of eNOS in the cells exposed to a risk factor (eg, high level of ox-LDL) are essential for maintaining a high gradient of NO concentration between the endothelium and the adjacent tissues that allows for NO-dependent long-distance signaling. Maintaining the adequate NO gradient in vasculature makes its diffusion efficient enough to be reached by NO typical targets, such as vasodilation and the inhibition of platelets. It is also worth noting that the presence of l-NAME significantly blocked the CaI-stimulated release of both NO and O2·−, and completely obliterated the differences in the levels of the detected molecules between the cells treated with or without ox-LDL. These observations further confirmed that, in our experimental model, eNOS was an enzymatic source producing simultaneously both radicals and that the pretreatment of the cells with ox-LDL efficiently increased the extent of eNOS activity dysfunction.
Although eNOS targeting is an attractive approach for preventing and treating atherosclerosis and other cardiovascular disorders, the phenomenon of eNOS uncoupling hampers the attempts to assess the efficacy of pharmacological interventions in modulating endothelial function. eNOS must be regarded as both an NO- and an O2·−-producing enzyme, and, therefore, eNOS may have a dual effect on vascular function, depending on its functional state.7,10,11 Our results provide evidence that the direct action of MNA+ on the endothelium in the healthy normocholesterolemic and hypercholesterolemic subjects improves the endothelium-dependent vasorelaxing effect produced by the enhancement of NO bioavailability and the reduction of eNOS-dependent oxidative stress. MNA+ is capable of improving NO bioavailability within the endothelium by acting toward either the prevention of endothelial dysfunction (prevention of eNOS uncoupling) in normal cells or the restoration of endothelial function (restoration of eNOS coupling) in the cells exposed to a specific cardiovascular risk factor (high level of ox-LDL). This effectively makes it a highly promising compound worth further evaluation in the treatment of hypercholesterolemia and atherosclerosis.
Sources of Funding
This research was supported by the Ministry of Science and Higher Education/the National Science Centre, Poland, and the Foundation for Polish Science.
The online-only Data Supplement is available with this article at http://hyper.ahajournals.org/lookup/suppl/doi:10.1161/HYPERTENSIONAHA.111.183210/-/DC1.
- Received September 14, 2011.
- Revision received October 2, 2011.
- Accepted January 24, 2012.
- © 2012 American Heart Association, Inc.
- Forstermann U,
- Munzel T
- Kalinowski L,
- Dobrucki IT,
- Malinski T
- Kalinowski L,
- Matys T,
- Chabielska E,
- Buczko W,
- Malinski T