EXTRACELLULAR NICOTINAMIDE ADENINE DINUCLEOTIDE (NAD) ARRESTS ENDOTHELIAL CELL PROLIFERATION

Background: presence of NAD in the extracellular environment acts as an autocrine/paracrine signal in different cell types (including smooth myocytes, neurons, human hematopoietic stem cells, granulocytes), mediating inflammatory responses, cell proliferation and neurotransmitter release. Extracellular NAD (eNAD) can trigger cell responses through a CD38-mediated production of NAD-derived Ca2+-mobilizing second messengers, and/or through P2Y1/P2Y11 activation. Aims: the main aims are to investigate which stimuli can trigger eNAD release from endothelial cells (EC) and the effect of eNAD on EC functions, including cytokine release and proliferation. Methods: Human Umbilical Vein Endothelial Cells (HUVEC) were challenged with 20 ng/ml TNFa for 1 or 24 h, or with horizontal shaking as mechanical stress for 3 min, and the levels of released eNAD were determined with a cycling enzymatic assay. Intact cells were incubated with 0.3 mM NAD and NAD degradation to ADPR and/or Adenosine (Ado) were evaluated by HPLC. HUVEC were exposed to exogenously added eNAD and cell proliferation/viability were determined by SRB assay. Cytokine release was measured by commercially available ELISA kits. Changes in the intracellular Ca2+ concentrations were spectrophotometrically determined in FURA2-loaded cells. HUVEC were transduced with a lentiviral vector to overexpress CD38. Cell permeability to lucifer yellow (LY) was evaluated with a microplate reader upon EC incubation with LY for 10 min. Results: eNAD release from HUVEC was evaluated upon stimulation of cells with a pro-inflammatory stimulus or with mechanical stress. EC incubation with TNFa for 1 or 24 h increased eNAD levels in the medium (by 3 fold). Mechanical stress also determined a 2.5-fold increase in eNAD. Incubation of HUVEC with TNFa for 24 h determined a 4-fold increase in the expression of Connexin-43 (Cx43), the hemichannel allowing NAD transport through the plasma membrane. The influx of LY was indeed enhanced in cells incubated for 24 h with TNFa, confirming an increased permeability of the over-expressed Cx43. HUVEC did not exhibit any detectable CD38-mediated enzymatic conversion of NAD to ADPR. Instead, Ado was produced starting from NAD at a rate of 0.05 nmol/min/mg protein. Thus, HUVEC can release eNAD, which could act as an autocrine/paracrine stimulus. Given the critical importance of EC in pro-inflammatory processes, we investigated whether eNAD (known to trigger pro-inflammatory responses in neutrophils and monocytes) could stimulate the release of cytokines from HUVEC. Incubation of cells with eNAD (up to 100 M final concentration for 24 h) did not stimulate the release of IL6 and IL8, both released in pro-inflammatory conditions by HUVEC. Next, eNAD effect on cell viability/proliferation was investigated. eNAD decreased cell viability when added at concentrations in the low micromolar range (from 10 M). Addition of 100 M eNAD significantly decreased cell proliferation (approximately by 80%), as monitored for 24, 48 and 72 h. PI staining and FACS analysis, performed after 8 h incubation with eNAD, suggested that eNAD slowed the cell cycle. In the attempt to elucidate the mechanism of eNAD effect on cell proliferation, HUVEC were incubated with the isomer aNAD, which is not a substrate for CD38 or PARP/ARTs, while being an activator of the P2Y11 receptor. aNAD, added at 100 uM, decreased cell proliferation as well as bNAD. This data confirms that a CD38-mediated conversion of eNAD is not necessary to obtain the eNAD-induced decrease in cell proliferation. However, the use of selective P2Y11 inhibitors (NF157 and NF340) did not rescue the eNAD-induced proliferation arrest. In addition, eNAD did not elicit any Ca2+ increase in HUVEC, as it would be expected in the case of activation of purinergic receptors. Thus, P2Y11 seems not to mediate the eNAD-induced decrease in cell proliferation. Ado affected cell viability when added in the micromolar range: thus, the NAD-derived production of Ado may represent a possible mechanism to explain eNAD effect. This hypothesis may be in line with the observation that HUVEC over-expressing CD38 (subtracting NAD to the generation of Ado from NAD) are less sensitive to the eNAD-induced cell proliferation arrest. Conclusions: eNAD may be autocrinally released by stimulated EC, or may derive from lysed cells in case of vascular injury, or may be released by other cell types in inflammatory conditions. Our results suggest that eNAD reduces EC proliferation, thus exhibiting an anti-angiogenic effect, possibly in line with the protective role, reported by others, exerted by eNAD against the EC barrier disruption induced by pro-inflammatory stimuli.