.
Posted on May 11th, 2020 by MikeDC
O P E N A C C E S S S O U R C E : ScienceDirect
Abstract
Background
Nicotinamide mononucleotide (NMN) and nicotinamide riboside (NR) are effective substrates for NAD synthesis, which may act as vasoprotective agents. Here we characterize the effects of NMN and NR on endothelial inflammation and dysfunction and test the involvement of CD73 in these effects.
Materials and methods
The effect of NMN and NR on IL1β or TNFα –induced endothelial inflammation (ICAM-1 and vWF expression), intracellular NAD concentration and NAD-related enzyme expression (NAMPT, CD38, CD73), were studied in HAECs. The effect of NMN and NR on angiotensin II-induced impairment of endothelium-dependent vasodilation was analyzed in murine aortic rings. The involvement of CD73 in NMN and NR effects was tested using CD73 inhibitor- AOPCP, or CD73-/- mice.
Results
24h-incubation with NMN and NR induced anti-inflammatory effects in HAEC stimulated by IL1β or TNFα, as evidenced by a reduction in ICAM-1 and vWF expression. Effects of exogenous NMN but not NR was abrogated in the presence of AOPCP, that efficiently inhibited extracellular endothelial conversion of NMN to NR, without a significant effect on the metabolism of NMN to NA. Surprisingly, intracellular NAD concentration increased in HAEC stimulated by IL1β or TNFα and this effect was associated with upregulation of NAMPT and CD73, whereas changes in CD38 expression were less pronounced. NMN and NR further increased NAD in IL1β-stimulated HAECs and AOPCP diminished NMN-induced increase in NAD, without an effect on NR-induced response. In ex vivo aortic rings stimulated with angiotensin II for 24 hours, NO-dependent vasorelaxation induced by Acetylcholine was impaired. NMN and NR, both prevented Ang II-induced endothelial dysfunction in the aorta. In aortic rings taken from CD73-/- mice NMN effect was lost, whereas NR effect was preserved.
Conclusion
NMN and NR modulate intracellular NAD content in endothelium, inhibit endothelial inflammation and improve NO-dependent function by CD73-dependent and independent pathways, respectively. Extracellular conversion of NMN to NR by CD73 localized in the luminal surface of endothelial cells represent important vasoprotective mechanisms to maintain intracellular NAD.
1. Introduction
Nicotinamide riboside (NR) and nicotinamide mononucleotide (NMN) have drawn attention as alternative nicotinamide adenine dinucleotide (NAD) substrates, devoid of side effects for nicotinic acid (NicA), such as “flushing” or hepatotoxicity and side effects of nicotinamide (NA), including sirtuin inhibition. Both NAD substrates, NR and NMN were proposed to be used in sports nutrition as good dietary supplements [78], [9], [77], [8] and display numerous beneficial effects in various settings, but their bioavailability and pathways of metabolism towards NAD differs.
NR, detectable in cow milk, milk-derived products and in natural products containing yeast [72], [3] has a good bio-availability and intracellularly is metabolized via NRK1 and NRK2 to NMN, a major precursor of NAD. NR was shown to be effective in restoring the NAD pool both in mice and humans [61]. Numerous studies on NR showed a significant impact of this substrate on NAD content, bioenergetics, and improved regenerative capabilities in various rodent models of disease. For example, NR treatment resulted in increased NAD concentration in a mouse model of respiratory chain III complex deficiency [52], improved liver regeneration [46] and restored NAD content in mouse skeletal muscle myotubes [86], [17]. NR treatment also enhanced oxidative metabolism and prevented weight gain in a mouse model of diet-induced obesity [76], [7]. Moreover, NR treatment increased NAD content in the cerebral cortex, thus attenuated cognitive deterioration in a mouse model of Alzheimer’s disease [21]. NR was also effective in heart failure, as NR-supplemented diet administrated to murine models of dilated cardiomyopathy or pressure overload-induced heart failure restored myocardial NAD levels and improved impaired cardiac function [15].
In contrast to NR, NMN has a worse bioavailability, as extracellular NMN is unable to pass the endothelial membrane without prior dephosphorylation by CD73 (ecto-5’-nucleotidase) to NR [93], [24] or prior to metabolism to nicotinamide by extracellular CD38 [94], [25], [97], [28]). Extracellular nicotinamide in the presence of phosphoribosyl-1-pyrophosphate (PRPP) could be also converted NMN by visfatin as reviewed recently [92], [23], [65]. Interestingly, in cardiomyocytes, it was demonstrated that connexin 43 (Cx43) channels are permeable to extracellular NAD [73], [4] suggesting that intracellular transport of NAD and NMN may be cell-type dependent and reliant on various transporters. Intracellularly, NA could be converted by nicotinamide phosphoribosyltransferase (NAMPT) to endogenous NMN in three-step Preiss-Handler pathway (Pei Wang, Wen-Lin Li, 2016; [89], [20]; Imai & Guarente, 2014), or methylated by nicotinamide N-methyltransferase (NNMT) to 1-methylnicotinamide (MNA) [70], [1]. NR is phosphorylated by nicotinamide riboside kinases (NRK1, NRK2) to endogenous NMN [53], [86], [17]. NMN is subsequently transformed to NAD by nicotinamide mononucleotide adenylyltransferases (NMNAT1-3). Apart from involvement in redox reactions, NAD is also substrate for sirtuins (SIRT), poly-ADP-ribose polymerases (PARP) and other NAD-dependent enzymes resulting in release of endogenous NA.
In numerous studies, NMN unequivocally afforded NAD-dependent beneficial effects. For example, NMN improved muscular contractile function in mouse age-related models of muscle dysfunction [86], [17], [45], restored cardiac NAD content in mouse model of ischemia-reperfusion [63], improved metabolic balance in type 2 diabetes mice [66], also improved NAD content and survival in rat models of hemorrhagic shock [55] and had a protective effect in β-amyloid oligomer-induced rat model of Alzheimer’s disease [62]. In some previous reports, the effects of NMN and NR were compared [49], [86], [17], but in most of these studies, either NMN or NR was characterized. Still, the role of ecto-enzymes CD73 and CD38 in NMN-induced effects has not been fully characterized, so it is not clear whether NMN-triggered beneficial effects are NR- or NA-dependent and what metabolic enzymes are involved.
Despite numerous studies on the beneficial effects of NR and NMN in various models, there is still a paucity of data as regards the effects of NMN and NR on endothelial function. NMN treatment had a beneficial effect in various mouse models of age-related vascular pathologies in line with the gradual fall in NAD content in aging [35], [82], [13]. These studies demonstrated that NMN restored endothelium-dependent vascular function and mitigated oxidative stress in age-related model [51], rescued angiogenic capacity in aged cerebrovascular endothelial cells [35] and restored fenestration-like phenotype of liver sinusoidal endothelial cells (LSECs) isolated from old mice [99], [30]. NR was also shown to be effective to improve vascular function. NR improved endothelium-dependent relaxation of isolated rat mesenteric arteries in ischaemia-reperfusion model [60]. Beneficial endothelial effects of NR was also shown in a mouse model of endotoxaemia, in which model NR restored NAD contents in lung and heart as well as decreased ROS production and apoptosis in isolated endothelial cells [98], [29].
In the present work, we aimed to characterize the endothelial profile of action of NMN in comparison to NR in cellular and vascular models of endothelial inflammation, with particular attention to the involvement of extracellular conversion of NMN in these effects. Our research demonstrated, that both NMN and NR modulated intracellular NAD content in the endothelium, inhibited endothelial inflammation and improved NO-dependent function. The important finding of this work was to show that NMN effects on endothelium were mediated by CD73-dependent conversion of NMN to NR.
2. Materials and methods
.../...
3. Results
3.1. Effect of CD73 inhibition by AOPCP on the extracellular conversion of NMN in Eahy.926 and HAEC cells
In Eahy.926 cells exposed to the increasing concentration of exogenous NMN, the release of NR and NA was increased as measured by HPLC assay (fig.1A and 1B). The Michaelis constant and Vmax for NMN→NR reaction was 1.37 mM and 1.138 nmol/ml/min, respectively, while for NMN→NR reaction: 2.29 mM and 0.583 nmol/ml/min, respectively. AOPCP (CD73 inhibitor) incubated for 2h at a concentration of 50 µM effectively diminished adenosine production from AMP by CD73 (fig.1C) as well as NR production from NMN, but had almost no effect on NA production from NMN (fig.1D). In HAECs AOPCP (incubated for 24 hours) also effectively inhibited conversion of NMN to NR as measured by LC/MS/MS assay (fig.1F), resulting in very low concentration of extracellular NR, as compared with NMN-treated group not pretreated with AOPCP (33.54 nmol/g of prot. vs 3355 nmol/g of prot, respectively, p≤0.01). NA production was only marginally affected by AOPCP (1147 nmol/g of prot. in NMN/AOPCP-treated group vs 1357 nmol/g of prot in NMN-treated group; p≤0.05).
Fig. 1. Effects of CD73 inhibition by AOPCP on extracellular metabolism of NMN in Eahy.926 (A-E) and HAEC cells (F, G). Michaelis constant and Vmax of extracellular nicotinamide riboside/adenosine (A) and nicotinamide production (B) by from exogenous nicotinamide mononucleotide and/or AMP. The rates of adenosine production from adenosine monophosphate (AMP) ©, nicotinamide riboside production from nicotinamide monophosphate (NMN) (D) and nicotinamide production from NMN (E) on the surface of EA.hy926 cells. The concentration of extracellular nicotinamide riboside (F) and nicotinamide (G) released by HAEC cells, measured after 24h-preincubation with CD73 inhibitor AOPCP and 60 min-incubation with nicotinamide mononucleotide (100µM). n=6, * p≤0.05, ** p≤0.01, *** p≤0.001.
3.2. Effects of NMN and NR on von Willebrand Factor and ICAM-1 expression in IL1β- and TNFα-stimulated HAEC cells; involvement of CD73
IL-1β induced the upregulation of vWF and ICAM-1 in HAECs, and NMN prevented the pro-inflammatory effects of IL-1β. The anti-inflammatory effect of NMN was lost in the presence of CD73 inhibitor AOPCP (50 µM). HAECs treated only with AOPCP displayed increased expression of vWF, as compared with untreated control HAECs (p≤0.05). IL1β–induced upregulation of vWF and ICAM1 expression was also reduced in the presence of NR but the anti-inflammatory effect of NR was not modified by AOPCP (fig.2).
Fig. 2. Effects of NMN and NR on IL1β –induced increase in vWF and ICAM-1 expression in HAECs. Expression of vWF in NR and NMN- and NR-treated HAEC cells after 24h-stimulation with IL1β in the presence or absence of CD73 inhibitor AOPCP, shown on representative images (A); expression of ICAM1 in NR- and NMN-treated HAEC cells after 24h-stimulation with IL1β and CD73 inhibitor AOPCP, demonstrated on representative images (B); n=6, * p≤0.05, ** p≤0.01; white scale bar represents 25 µm.
3.3. Effects of NMN and NR on intracellular NAD concentration in HAECs after stimulation with IL1β; effects of CD73 inhibition
NMN or NR raised intracellular NAD content in basal non-stimulated HAECs (fig.4). NMN-triggered NAD increase was a CD73-dependent response, since, in AOPCP-treated HAECs, NMN supplementation did not increase significantly NAD content. In contrast to NMN, NR-induced raise of NAD was not modified by AOPCP. Stimulation with IL1β was not linked to the significant fall in intracellular NAD content, rather an increased was noted CD73 inhibition by AOPCP prevented the rise in NAD induced by NMN in IL1β-stimulated HAECs, however effects of NR on NAD content in this experimental setup was not modified by AOPCP.
Fig. 4. Effects of NR and NMN on NAD content in HAEC cells, in the presence and absence of CD73 inhibition by AOPCP. Effect of NR (A) and NMN (B) on intracellular NAD after 24h-stimulation with IL1β-in the presence or the absence of CD73 inhibitor AOPCP 50µM; n=5, * p≤0.05, ** p≤0.01, ns- not statistically significant
3.4. Changes in NAMPT, CD38 and CD73 expression in IL1β-stimulated HAECs
To explain the increase of NAD content in IL1β -stimulated HAECs we analyzed the expression of NAMPT and two ectoenzymes; CD73 and CD38 in this experimental setting by immunocytochemistry and Western Blot. As shown in fig.5A, a specific immunofluorescence of NAMPT, the main cytosolic enzyme involved in NAD synthesis was upregulated (p≤0.01) in HAECs stimulated by IL1β. CD73-specific immunofluorescence was higher in IL1β- treated HAECs, as compared with untreated HAECs (fig.5B). Similarly, CD73 and NAMPT expression in HAEC stimulated by IL1β were increased as assessed by Western Blot (fig.5D). Supplementation with NMN or NR resulted in a decrease of upregulated NAMPT and CD73 expression in IL1β-stimulated HAECs, which was also confirmed by WB assay. Stimulation with IL1β had no significant effect on CD38-specific immunofluorescence (fig.5C), while WB analysis shown a minor upregulation of CD38 in IL1β-treated cells and downregulation after co-incubation with NMN or NR. This data suggests the compensatory upregulation of NAMPT, CD73 in response to inflammatory stimulus, might contribute to the preservation of NAD pool in IL1β-stimulated HAECs and NMN or NR treatment resulting in the anti-inflammatory effects, reverted compensatory upregulation of NAMPT and CD73 expression in IL1β-stimulated HAECs.
Fig. 5. Upregulation of NAMPT, CD73 and CD38 in HAEC after 24h-stimulation with IL1β, assessed by immunofluorescent staining and Western Blot assay; results of NAMPT immunofluorescent imaging in IL1β-stimulated cells in the presence or absence of NMN or NR (A); expression of CD73 (5’ectonucleotidase) in IL1β-stimulated cells in the absence or presence of NMN or NR assessed by immunofluorescent imaging (B); results of CD38 imaging after incubation with IL1β in the presence or absence of NMN or NR ©; results of Western Blot analysis of NAMPT, CD73 and CD38 expression after stimulation with IL1β, in a presence or absence of NMN or NR (D); n=6, * p≤0.05, ** p≤0.01; white scale bar represents 25 µm.
.../...
.
