Visfatin/eNampt induces endothelial dysfunction in vivo: a role for Toll-Like Receptor 4 and NLRP3 inflammasome

Visfatin/extracellular-nicotinamide-phosphoribosyltranferase-(eNampt) is a multifaceted adipokine enhanced in type-2-diabetes and obesity. Visfatin/eNampt cause in vitro endothelial dysfunction and vascular inflammation, although whether the same effects are achieved in vivo is unknown. Toll-like receptor-4 (TLR4), a main surface pattern recognition receptor of innate immune system is a potential target for visfatin/eNampt. We studied its capacity to generate vascular dysfunction in vivo, focusing on TLR4 role and downstream activation of nod-like-receptor-protein-3 (NLRP3)-inflammasome. 4 month-old C57BL/6 mice were exposed to 7 days infusion of visfatin/eNampt, alone or together with FK 866 (Nampt enzymatic inhibitor), CLI 095 (TLR4 blocker), MCC 950 (NLRP3-inflammasome inhibitor), or anakinra (interleukin(IL)-1-receptor antagonist). Endothelial dysfunction was tested in isolated microvessels. In human umbilical endothelial cells (HUVEC), proteins related to the NLRP3-inflammasome phosphorylated p-65, NLRP3, caspase-1, pro-IL-1β, and mature IL-1β were determined by Western blot, while the inflammasome related apoptosis-associated speck-like protein containing a caspase recruitment domain (ASC-specks) was studied by immunofluorescence. Impaired endothelium-dependent relaxations were observed in isolated mesenteric microvessels from visfatin/eNampt-infused mice. This effect was attenuated by co-treatment with FK 866 or CLI 095, supporting a role for Nampt enzymatic activity and TLR4 activation. Moreover, cultured HUVEC exposed to visfatin/eNampt showed higher expression and activation of NLRP3-inflammasome. Again, this effect relied on Nampt enzymatic activity and TLR4 activation, and it was abrogated by the inflammasome assembly blockade with MCC 950. The endothelial dysfunction evoked by visfatin/eNampt infusion in vivo was also sensitive to both MCC 950 and anakinra treatments, suggesting that the NLRP3-inflammasome-driven tissular release of IL-1β is the final mediator of endothelial damage. We conclude that Visfatin/eNampt produces in vivo vascular dysfunction in mice by a Nampt-dependent TLR4-mediated pathway, involving NLRP3-inflammasome and paracrine IL-1β. Thus, those targets may become therapeutic strategies for attenuating the adipokine-mediated vascular dysfunction associated to obesity and/or type-2-diabetes.

Additionally, a group of control untreated animals were used for analysing the ex vivo effects of the drugs. To sacrifice, the animals were briefly exposed to a chamber filled with carbon dioxide until they fell unconscious and then immediately killed by cervical dislocation. On day 0 and 6, weight, plasma glucose, and mean arterial pressure were measured, leading to minor changes: a small weight reduction was observed in CLI 095-treated mice while significant lower plasma glucose levels were obtained in IL-1-treated animals (Table S1).

Biochemical data
Blood plasma samples were collected by venipuncture in a BD Vacutainer® with heparin (BD, Franklin Lakes, NJ, USA) from cava vein and stored at − 80°C until assay was performed. Serum was prepared according to the manufacturer's recommendations of inversion and centrifugation and aliquoted and stored at -80°C until use in experiments.

Drug effects on vascular tone of mesenteric microvessels
For reactivity experiments, the mesentery was removed, placed in a Petri dish containing Krebs-Henseleit solution (KHS) at 4°C. The third branch mesenteric arteries were dissected (mean internal diameter ranged between 150-400 µm, with non-significant differences observed among the different groups of mice). The arteries were dissected cleaned free of fat and connective tissue under a light microscope and mounted as ring preparations on a small vessel myograph [11] capable of measuring isometric tension. Arteries were bathed in KHS at 37°C continuously bubbled with a 95% O 2 -5% CO 2 mixture, which gives a pH of 7.4 and their passive tension and internal circumference were determined. The arteries were subjected to optimal tension (90% of the tension equivalent to a intramural pressure of 100 mm Hg. After 30 min of equilibration, the vessels were exposed to 125 mmol/L K + (KKHS, equimolar substitution of KCl for NaCl in KHS) for 2 min to check their functional integrity. Segments failing to produce a maximum active tension equivalent to a pressure of 100 mmHg on the final contraction were rejected [11].
The bath was then washed three times with KHS and further 60-120 min washout period was allowed before the arteries were contracted with the concentration of noradrenaline (NA; 3 µmol/L) required to produce approximately 80% of the maximum response to KKHS.

Cell Culture
Human umbilical vein endothelial cells (HUVEC) were isolated from umbilical cords, as previously described [7]. Cells were cultured in M199 medium supplemented with 20% foetal calf serum (FCS), 25 μg/mL endothelial cell growth supplement (ECGS), 100 μg/mL heparin and antibiotics (100 U/ml penicillin, 100 µg/ml streptomycin and 2.5 µg/ml amphotericin B) at 37°C in a humidified atmosphere with 5% CO2. Cells at passages 1-5 were incubated for the indicated time periods with the different test compounds in M199 medium supplemented with 10% FCS, ECGS and antibiotics. All the procedures were reviewed and approved by the ethics committee of Universidad Autónoma of Madrid and Hospital Universitario La Paz, respectively, and written informed consent was obtained from all cord donors.

Western blot analysis
At the end of the treatment periods, the levels of selected proteins were detected by Western blot in HUVEC or in aortic and renal homogenates, as previously described [7]. Primary antibodies were used against phospho-p65 (P-p65), total p65, NLRP3, pro-caspase-1 and cleaved(cle)caspase-1 forms, IL-1β, or pro-IL-1β (Cell Signalling, Adipogen; Novus Biologicals; R&D systems, respectively, at 1/1,000 dilution) and anti β-actin primary antibody (dilution 1/10,000; Sigma-Aldrich) to ensure equal loading, followed by incubation with corresponding specific horseradish peroxidase-conjugated secondary antibodies (Bio-Rad or Bethyl; 1:10,000) Immunoreactive bands were detected using an ECL detection kit (GE Healthcare) and quantified by densitometry using the NIH software Image J.

Indirect immunofluorescence
The activation of the inflammasome requires ASC proteins to assemble into a large toroidal protein complex, which is termed "speck" [23]. ASC-specks were visualized in HUVEC by indirect immunofluorescence as previously described [7]. A primary polyclonal antibody against ASC (dilution 1:250; Molecular Probes) was used, followed by incubation with an appropriate Alexa 546-conjugated secondary antibody (dilution 1:100; Molecular Probes). Nuclei were counterstained with DAPI (5 µg/mL, Invitrogen) and cells were observed with a confocal microscopy (TCS SPE, Leica, Wetzlar, Germany). The percentage of specks was calculated as the percentage of cells displaying specks versus the total number of cell per field. Specks were first counted manually with a fluorescence microscope (Eclipse TE300; Nikon, Tokyo, Japan). In every preparation, we initially selected a central field and afterwards, following a counterclock radial pattern, eighth radius were traced and two fields were explored in each of them (total measurements, 17 fields per preparation at 100X). Representative images (63X) were obtained from every preparation with a confocal microscope (TCS SPE, Leica, Wetzlar, Germany).

Histological studies
Serial paraffin sections (4 μm) of half-height sliced myocardium or kidneys were fixed on slides and stained with Haematoxylin/Eosin (H/E). In parallel, Masson trichrome was used to detect extra-cellular matrix (ECM) deposition by sequent addition of Bouin's, Weigert's and Biebrich solutions (Bio-Optica, Milan, Italy) on paraffin sections (4 μm) of all myocardia and kidney samples. Interstitial, perivascular and replacement fibrosis were quantified together on five fields of each organ preparation using the Metamorph software. Photographs were taken at 40x magnification under optical microscopy (Eclipse TE300; Nikon, Tokyo, Japan).

Supplementary legends
Additional file 1: Table S1. Vascular tone achieved with noradrenaline prior to relaxation curves and pEC50 values for ACh or SNP in isolated mesenteric microvessels from C57BL/6 mice infused for 7 days with visfatin/eNampt or IL-1.
Additional file 2: Table S2. Weight, plasmatic glucose, and mean arterial pressure in C57BL/6 mice infused for 7 days by osmotic mini-pumps.
Additional file 3: Table S3. Vascular tone achieved with noradrenaline prior to relaxation curves and pEC50 values for ACh or SNP in isolated mesenteric microvessels from untreated C57BL/6 mice.
Additional file 4: Figure S1. Schematic representation of the designed experimental groups of C57BL/6 mice infused with visfatin/eNampt or IL-1 through subcutaneously implanted osmotic minipumps and the specified durgs.  or after receiving 5 ng/mL IL-1, 100 µg/mL anakinra, or both IL-1 plus anakinra. (B) Isolated mice mesenteric microvessels were obtained from animals infused during 7 days, through subcutaneous osmotic mini-pumps, with saline solution or IL-1 (12 µg/kg/day), and receiving also intraperitoneal administration of anakinra (100 mg/kg/day) or analogous amounts of saline during the last 3 days before sacrifice. The isolated vessels were pre-contracted with NA and submitted to cumulative concentrations of ACh. The curves (mean±SEM) are expressed as the percentage of the previous NA-evoked contraction, which is indicated in the Tables S2 and S3, as well as the respective pEC50 values. For every curve, 7 to 11 segments were used, obtained from 4 to 8 different animals. *p<0.05 vs respective control. #p<0.05 vs respective IL-1.  Mesenteric microvessels from untreated C57BL/6 mice were isolated and mounted in a myograph. After contracting the vessels with 3 µmol/L noradrenaline (NA), endotheliumdependent relaxations were studied by adding ex vivo cumulative concentrations of acetylcholine (ACh, 10 nmol/L to 10 μmol/L) into the organ bath in the absence of any other treatment or with the indicated drugs also administered ex vivo. Using a similar protocol, the endotheliumindependent relaxations to sodium nitroprusside (SNP; 1 nmol/L to 10 nmol/L) were assessed. The contractile responses evoked by NA (in mNewtons, mN), as well as the respective pEC50 values for ACh or SNP are indicated (mean±SEM). *p<0.05 vs control; +p<0.05 vs respective visfatin/eNampt or NMN; #p<0.05 vs IL-1. Visfatin/eNampt, FK 866, visfatin/eNampt plus FK 866, and IL-1 were infused for 7 days in C57BL/6 mice through osmotic minipumps. a These animals received i.p. CLI 095 (days 1 to 6), MCC 950 (days 2, 4, and 6), or anakinra (days 4, 5, and 6) before sacrifice. Afterwards, mesenteric microvessels were isolated and mounted in a myograph. The vessels were then precontracted with 3 µmol/L noradrenalilne (NA), and the endothelium-dependent relaxations were studied by adding cumulative concentrations of acetylcholine (ACh, 10 nmol/L to 10 μmol/L) into the organ bath. Using a similar protocol, the endothelium-independent relaxations to sodium nitroprusside (SNP; 1 nmol/L to 10 nmol/L) were studied. The contractile responses by NA (in mNewtons, mN), as well as the respective pEC50 values for ACh or SNP are indicated (mean±SEM). *p<0.05 vs saline solution; +p<0.05 vs visfatin/eNampt; #p<0.05 vs IL-1.