Background: High-density lipoproteins (HDLs) can exert anti-atherogenic effects. On top of removing excess cho - lesterol through reverse cholesterol transport, HDLs play beneficial actions on endothelial function and integrity. In particular, HDLs are strong determinant of endothelial progenitor cells (EPCs) number and function. To gain further insights into such an effect we characterized in vitro functionality of circulating “early” EPCs obtained from 60 type 2 diabetes individuals with low HDL-cholesterol (HDL-C) and 59 with high HDL-C levels. Methods: After an overnight fast, venous blood was drawn in EDTA tubes and processed within 2-h from sampling. Peripheral blood mononuclear cells were isolated and plated on fibronectin coated culture dishes; after 3 days cul- ture, adherent cells positive for Dil-ac-LDL/Lectin dual fluorescent staining were identified as monocytic angiogenic cells (MACs). After 5–7 days culture in EBM-2 medium, adherent cells were evaluated for viability/proliferation (MTT assay), senescence (beta-galactosidase activity detection), migration (modified Boyden chamber using VEGF as chem- oattractant), adhesion capacity (on fibronectin-coated culture dishes) and ROS production (ROS-sensitive fluorescent probe CM-H DCFDA). Results: MACs obtained from diabetic individuals with high HDL-C had 23% higher viability compared to low HDL-C (111.6 ± 32.7% vs. 90.5 ± 28.6% optical density; p = 0.002). H O exposure impaired MACs viability to a similar extent 2 2 in both groups (109.2 ± 31.7% vs. 74.5 ± 40.8% in high HDL-C, p < 0.0001; 88.3 ± 25.5% vs. 72.3 ± 22.5% in low-HDL, p = 0.004). MACs senescence was comparable in the two groups (102.7 ± 29.8% vs. 99.2 ± 27.8%; p = 0.703) and was only slightly modified by exposure to H O . There was no difference in the MACs migration capacity between 2 2 the two groups (91.3 ± 34.2% vs. 108.7 ± 39.5%; p = 0.111), as well as in MACs adhesion capacity (105.2 ± 32.7% vs. 94.1 ± 26.1%; p = 0.223). Finally, ROS production was slightly thought not significantly higher in MACs from type 2 dia- betes individuals with low- than high-HDL. After stratification of HDL-C levels into quartiles, viability (p < 0.0001) and adhesion (p = 0.044) were higher in Q4 than in Q1–Q3. In logistic regression analysis, HDL-C was correlated to MACs viability and adhesion independently of HbA1c or BMI, respectively. *Correspondence: firstname.lastname@example.org Daniela Lucchesi and Simona Georgiana Popa contributed equally to this work Section of Diabetes and Metabolic Disease, Department of Clinical and Experimental Medicine, University of Pisa and Azienda Ospedaliero- Universitaria Pisana, Via Paradisa, 2, 56124 Pisa, Italy Full list of author information is available at the end of the article © The Author(s) 2018. This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creat iveco mmons .org/licen ses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creat iveco mmons .org/ publi cdoma in/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Lucchesi et al. Cardiovasc Diabetol (2018) 17:78 Page 2 of 13 Conclusions: Our data suggest that in type 2 diabetes subjects, HDL-cholesterol is an independent determinant of circulating MACs functional capacities—mainly viability, to a lesser extent adhesion—likely contributing also through this mechanism to cardiovascular protection even in type 2 diabetes. Keywords: Type 2 diabetes mellitus, Endothelial progenitor cells, Monocytic angiogenic cells, High-density lipoprotein cholesterol, Endothelial function Background This already complex pattern of interaction between Epidemiological and clinical studies indicate that plasma HDL particles and the endothelium becomes even more levels of high-density lipoproteins (HDLs) cholesterol complicated in the diabetic condition [3, 15, 16] as both (HDL-C) are associated with lower risk of coronary HDLs  and EPCs [18, 19] from diabetic patients artery disease. Indeed, HDL particles are known to exert have been found to be dysfunctional. Changes in HDLs a broad spectrum of anti-atherogenic properties : metabolism related to insulin resistance, glycation and they remove cholesterol excess through reverse choles- depletion of apolipoprotein A-1 (apoA-1), modifica - terol transport from peripheral tissue to the liver, inhibit tions of other HDL-associated proteins such as paraoxo- lipid oxidation, exert anti-inflammatory, anti-throm - nase-1 (PON-1) or oxidation of components of the HDL botic, anti-proteolytic and anti-apoptotic effects, inhibit particles, and alterations of HDL proteome induced by intracellular oxidative stress, and restore endothelial chronic inflammation, all contribute to HDL dysfunction dysfunction . HDLs also contribute to nitric oxide bio- in diabetes . Compared with healthy subjects, HDLs availability to ensure vasodilation  and facilitate repair from individuals with type 2 diabetes show impaired of endothelial injuries [1, 2]. in vitro endothelial-protective effects in the aortic ring Endothelial repair is largely dependent on recruitment segments from mice , while infusion of rHDLs in type of circulating endothelial progenitor cells (EPCs) in the 2 diabetes patients restored serotonin induced vasodila- areas of intimal injury and activation of quiescent tissue- tion . In particular, HDLs from healthy, but not from resident endothelial cells or resident EPCs via the release diabetic subjects, promoted in vivo endothelial repair of paracrine factors . Resident endothelial cells/EPCs by early EPCs obtained from patients with diabetes in a and endothelial cell-derived microparticles (EMPs) can model of carotid injury in mice . Furthermore, in dia- provide direct sources of paracrine factors with angio- betic subjects the number of circulating EPCs is reduced genic and angio-protective properties [2, 4]. and their function altered even in the absence of diabetic HDLs exert a potent effect on the number and function complications [18, 19]. of circulating EPCs [5–8]. In vitro, human HDLs enhance Within this scenario, we now report results of a study differentiation of mononuclear cells into early EPCs, that evaluated the relationship between HDL-cholesterol inhibit their apoptosis [9, 10], increase their migratory levels and ex vivo functional properties of monocytic capacity, and propensity to adhesion . In a model of angiogenic cells (MACs), originally defined as “early” −/− apolipoprotein E-deficient (apoE ) mice, intravenous EPCs, isolated from individuals with type 2 diabetes. infusion of reconstituted HDLs (rHDLs) more than dou- bled the number of EPCs engrafted in the aortic endothe- Methods lium . In EPCs isolated from the bone marrow of Patients characteristics and study design hypercholesterolemic rats, HDLs promoted EPCs pro- A total of 119 men and women with type 2 diabetes for liferation, migration, and tube formation . In a low- more than 12 months and on stable anti-diabetic treat- −/− density lipoprotein (LDL) receptor deficient (LDLr ) ment regimen (oral anti-diabetic medications and/ mice, lipid lowering or HDLs resulted in a two-fold or long acting or short acting insulin) for the prior increase of circulating EPCs and improved function as 3 months were consecutively recruited at the Diabetes assessed by ex vivo EPCs migration and adhesion . Unit of the Department of Clinical and Experimental Finally, in a recent study, impaired viability of late out- Medicine of the University of Pisa. Pregnant or breast- growth EPCs (generated from human peripheral mono- feeding women, individuals of non European descent, nuclear cells, PBMCs) induced by oxidized LDL (oxLDL) subjects with type 1 diabetes, those with advanced, was reversed by HDLs in a dose dependent manner. Of sight-threatening diabetic retinopathy or chronic kidney note, in the absence of oxLDL, low HDL concentration disease (CKD) stages ≥ 3b, as well those with poor gly- enhanced EPCs tube formation, while moderate to high cemic control, defined as HbA1c ≥ 9.0% as measured at concentrations paradoxically enhanced EPCs senescence the enrollment visit, were excluded. Further exclusion and impaired tube formation . criteria were body mass index (BMI) ≥ 35 kg/m , severe Lucchesi et al. Cardiovasc Diabetol (2018) 17:78 Page 3 of 13 uncontrolled hypertension defined as systolic blood pres - all documented major acute CVD events and revasculari- sure (BP) ≥ 180 mmHg and/or diastolic BP ≥ 110 mmHg, zation procedures. Peripheral vascular disease was also sign of acute illness or infection, significant hepatic dis - assessed by search of femoral and foot pulses, and meas- ease, chronic inflammatory disease, immunological and urement of ankle/brachial pressure ratio . myeloproliferative diseases or other malignancy, and any Patients were divided in subjects with low [≤ 40/50 mg/ major cardiovascular event within 3 months of entering dl (≤ 1.034/1.293 mmol/l) in men/women] and high in the study. All patients underwent a structured inter- HDL-C levels [> 40/50 mg/dl (> 1.034/1.293 mmol/l) in view in order to collect information about the onset of men/women]. The main characteristics of the two study diabetes, its duration, smoking habits, current anti-dia- groups are given in Table 1. In all these subjects, a 50 ml betic treatment, and BP- and lipid-lowering therapies fasting blood sample was drawn in EDTA-containing . Body weight and height were assessed and BMI cal- tubes and processed within 2 h from collection for cul- culated; waist circumference was measured at midway ture and assessment of MACs functional properties (see between the costal margins and the iliac crests. BP was below). measured after 5-min rest while seated and the average The Ethics Committee of the University of Pisa of two consecutive measurements obtained about 5-min approved the study protocol and written informed con- apart was calculated. Hypertension was defined as sys - sent was obtained from all participants before any study tolic BP > 140 mmHg and/or diastolic BP > 80 mmHg procedure. and/or use of any antihypertensive medication. In all subjects a venous blood specimen was drawn after Cell culture an overnight fast for determination of HbA1c (high-per- MACs were cultured according to previously described formance liquid chromatography using DCCT-aligned techniques . PBMCs were isolated from blood of methods) , serum creatinine, glucose, total-, HDL-, each individual with type 2 diabetes by Biocoll (Biochrom and LDL-cholesterol (Friedewald formula), triacylglyc- AG; density = 1.077 g/ml) density-gradient centrifuga- erol, alanine aminotransferase (ALT), aspartate ami- tion. Total PBMCs were seeded on 2 µg/cm fibronectin ™ ® notransferase (AST), gamma-glutamyltransferase (GGT), coated culture dishes (BD Falcon ) or on Lab-Tek II uric acid, fibrinogen and blood cell count using standard chamber slides system (Sigma-Aldrich Ltd, Poole, Dor- laboratory methods. A first-voided urine sample was also set, UK) after red cell lyses. After isolation, cells were cul- collected for determination albumin (BNII; Dade Behring tured in endothelial basal medium (EBM-2, Lonza Sales Diagnostic, Marburg, Germany) and creatinine (modified AG, Basel, Switzerland) supplemented with EGM-2-MV- Jaffé reaction). Urinary samples with abnormal sediments SingleQuots containing human endothelial growth factor on routine analysis were excluded. Finally, all patients (EGF), hydrocortisone, insulin-like growth factor-1 (IGF- underwent a screening for diabetic complications as 1), fibroblast growth factor (FGF), vascular endothelium detailed elsewhere . Briefly, diabetic retinopathy was growth factor (VEGF), ascorbic acid, antibiotics and 5% assessed by retinal photography and its severity classi- fetal bovine serum (FBS, Lonza Sales AG). After 3 days fied according to the Global Diabetic Retinopathy Project of culture, non-adherent cells were discarded by wash- Group criteria ; patients with advanced, sight-threat- ing with phosphate buffered saline (PBS) and the culture ening retinopathy (severe non-proliferative, prolifera- medium was replenished daily. On day 5, adherent cells tive, maculopathy or blindness) were excluded. Based on displaying an elongated spindle-shaped morphology were ACR (albumin-to-creatinine ratio) values, UAE (urinary identified as MACs. albumin excretion) categories were defined as normoal - buminuria (< 30 mg/g, < 3.4 mg/mmol), microalbuminu- MACs characterization ria or “moderately increased albuminuria” (30–299 mg/g, MACs (“early” EPCs) were characterized for the uptake 3.4–34 mg/mmol) and macroalbuminuria or “severely of 1,10-dioctadecyl-3,3,3,3-tetramethylindocarbo-cya- increased albuminuria” (≥ 300 mg/g, ≥ 34 mg/mmol). nine-labeled acetylated low-density lipoprotein (DiI-ac- Estimated glomerular filtration rate (eGFR) was calcu - LDL) and for lectin binding. The staining was performed lated by the CKD-EPI (Chronic Kidney Disease Epi- incubating cells with 10 µg/ml of DiI-ac-LDL (Invitrogen, demiology Collaboration) equation  and subjects Life Technologies Ltd, Paislet, UK) for 2 h at 37 °C in dark with eGFR < 45 ml/min/1.73 m (CKD stages ≥ 3b) were conditions. Cells were fixed in 4% paraformaldehyde for excluded. Diabetic neuropathy was assessed by a vali- 30 min and counterstained with 1 mg/ml FITC-labelled dated questionnaire, by knee and ankle reflexes and by lectin from Ulex europaeus (Sigma-Aldrich Ltd) for 2 h measurement of vibration perception threshold (VPT) at 37 °C in dark conditions. Stained cells were observed . Finally, previous cardiovascular diseases (CVD) by a fluorescence microscope and double positive DiI-ac- were determined based on medical history by recording LDL/Lectin cells were identified as MACs. To evaluate Lucchesi et al. Cardiovasc Diabetol (2018) 17:78 Page 4 of 13 Table 1 The main anthropometric and clinical characteristics of the study cohort All subjects High HDL Low HDL p HDL > 40/50 mg/dl (M/F) HDL ≤ 40/50 mg/dl (M/F) N. (%) 119 59 (49.6%) 60 (50.4%) Age, years 63.6 ± 7.9 65.0 ± 8.4 62.1 ± 7.2 0.047 Duration of diabetes, years 11.8 ± 9.8 13.7 ± 10.7 9.8 ± 8.5 0.031 Sex (M/F), n (%) 69/50 (58/42) 25/34 (42/58) 44/16 (73/27) 0.001 28.8 ± 5.4 27.2 ± 5.8 30.4 ± 4.4 0.001 BMI, kg/m BMI categories (< 25, 25–30, > 27/54/38 (22.7/45.4/31.9) 22/27/10 (37.3/45.8/16.9) 5/27/28 (8.3/45.0/46.7) 0.0001 30 kg/m ), n (%) Waist circumference, cm 102.4 ± 13.3 97.6 ± 14.0 107.2 ± 10.7 0.0001 Systolic BP, mmHg 145.6 ± 17.1 145.2 ± 15.8 146.0 ± 18.5 0.813 Diastolic BP, mmHg 79.5 ± 9.1 77.1 ± 8.9 81.9 ± 8.6 0.004 Hypertension, n (%) 90 (75.6) 40 (67.8) 50 (83.3) 0.048 Fasting glucose, mg/dl (mmol/l) 146.4 ± 35.4 (8.13 ± 1.97) 147.1 ± 34.7 (8.17 ± 1.93) 145.6 ± 36.3 (8.09 ± 2.02) 0.814 HbA1c, % (mmol/mol) 7.27 ± 0.88 (56.0 ± 9.6) 7.24 ± 0.81 (55.6 ± 8.8) 7.31 ± 0.95 (56.3 ± 10.4) 0.683 Total cholesterol, mg/dl (mmol/l) 181.2 ± 33.4 (4.68 ± 0.86) 191.3 ± 34.2 (4.94 ± 0.89) 171.2 ± 29.6 (4.43 ± 0.77) 0.001 LDL cholesterol, mg/dl (mmol/l) 101.1 ± 27.8 (2.62 ± 0.72) 100.9 ± 29.0 (2.61 ± 0.75) 101.4 ± 26.8 (2.62 ± 0.69) 0.930 HDL cholesterol, mg/dl (mmol/l) 53.3 ± 42.0 (1.38 ± 0.58) 73.5 ± 13.0 (1.90 ± 0.34) 33.5 ± 5.1 (0.86 ± 0.13) – Non-HDL cholesterol, mg/dl 127.8 ± 31.9 (3.31 ± 0.83) 117.8 ± 31.6 (3.05 ± 0.82) 137.8 ± 29.3 (3.56 ± 0.76) 0.001 (mmol/l) Triacylglycerol, mg/dl [mmol/l] 124.0 (75.0–194.0) [1.40 81.0 (59.0–120.0) [0.92 (0.67– 185.5 (131.2–237.7) [2.10 0.0001 (0.85–2.19)] 1.36)] (1.48–2.69)] Creatinine, mg/dl (µmol/l) 0.90 ± 0.28 (79.6 ± 24.8) 0.80 ± 0.17 (70.8 ± 14.8) 1.00 ± 0.33 (88.4 ± 29.3) 0.0001 Uric acid, mg/dl (µmol/l) 5.35 ± 1.46 (318.2 ± 86.8) 4.83 ± 1.27 (287.2 ± 75.5) 5.88 ± 1.46 (349.7 ± 86.7) 0.0001 Albumin to creatinine ratio, mg/g 7.3 (3.9–18.8) 7.4 (4.6–13.7) 6.1 (3.4–36.4) 0.457 A/C ratio categories: < 30, 30–300, 95/20/4 (79.8/16.8/3.4) 50/7/2 (84.7/11.9/3.4) 45/13/2 (75.0/21.7/3.3) 0.358 > 300 mg/g; n (%) eGFR, CKD-EPI, ml/min/1.73 m 82.5 ± 18.3 85.8 ± 14.7 79.2 ± 20.9 0.052 AST, U/L 22.0 ± 14.9 20.8 ± 11.6 23.2 ± 17.7 0.377 ALT, U/L 23.7 ± 14.3 21.2 ± 12.6 26.1 ± 15.6 0.063 GGT, U/L 32.6 ± 29.4 33.9 ± 33.6 31.2 ± 24.9 0.624 Smoking habits: no smokers, ex- 63/33/23 (52.9/27.7/19.3) 39/16/4 (66.1/27.1/6.8) 24/17/19 (40.0/28.3/31.7) 0.001 smokers, current smokers, n (%) Glucose lowering treatments Metformin, n (%) 97 (81.5) 48 (81.4) 49 (81.7) 0.965 Sulphonilureas or glinides, n (%) 34 (28.6) 17 (28.8) 17 (28.3) 0.887 Thiazolidinediones, n (%) 6 (5.0) 4 (6.8) 2 (3.3) 0.390 DPP4 inhibitors, n (%) 36 (30.3) 15 (25.4) 21 (35.0) 0.256 GLP-1 receptor agonists, n (%) 11 (9.2) 3 (5.1) 8 (13.3) 0.120 Insulin, n (%) 35 (29.4) 19 (32.2) 16 (26.7) 0.507 BP-lowering treatments, n (%) 84 (70.6) 36 (61.0) 48 (80.0) 0.023 RAAS inhibitors, n (%) 70 (58.8) 32 (54.2) 38 (63.3) 0.313 Lipid-lowering treatments, n (%) 58 (48.7) 30 (50.8) 28 (46.7) 0.648 Non advanced retinopathy, n (%) 18 (15.1) 7 (11.9) 11 (18.3) 0.325 Major acute cardiovascular events 21 (17.6) 7 (11.9) 14 (23.3) 0.101 (MACE), n (%) the immunophenotype of MACs, adherent cells were (BioLegend), CD31-FITC (BD Biosciences), CD45-FITC detached with trypsin–EDTA and 5 × 10 cell/tube were (BD Biosciences) and CD42-PE (BD Biosciences) for incubated with anti-human CD34-PE (BD Biosciences), 30 min in dark conditions at 4 °C. Isotype control anti- CD133-PE (Miltenyi Biotec), VEGFR-2-Alexa Fluor 647 bodies were used to set baseline fluorescence levels. The Lucchesi et al. Cardiovasc Diabetol (2018) 17:78 Page 5 of 13 labeled cells were analyzed on a FACS-Calibur Instru- and exhibited double positivity for Ac-LDL and lectin ment (BD Biosciences), acquiring 2 × 10 events for each binding (UEA-1-FITC) as established by fluorescent analysis. The flow cytometric analysis was repeated six microscope analysis (Fig. 1a). MACs phenotype was con- times. After a 5 days culture under standard conditions, firmed by the expression of main endothelial cell surface MACs resulted in an adherent population consisting of markers: CD14 (99.06 ± 0.55%), CD31 (42.73 ± 25.24%), cells that showed elongated with a spindle-like shape CD34 (44.24 ± 34.74%), CD42 (1.56 ± 1.77%), CD45 UEA-1-FITC Dil-Ac-LDL Merge CD14 CD31 CD34CD42 99,06±0,55% 42,73±25,24% 44,24±34,74%1,56±1,77% CD45 CD133 VEGFR2 96,09±4,07% 9,27±6,93% 44,15±21,78% Fig. 1 a MACs phenotype characterization by double staining with DiI-Ac-LDL uptake (on the left; red) and lectin UEA-1-FITC binding (in the middle, green). Merged images showed DiI-Ac-LDL/lectin double-positive MACs (on the right, yellow) (magnification 20×). b FACS quantification of the cell surface markers in MACs. The picture shows results (expression of each surface marker; mean ± SD) typically obtained from six separate experiments. Isotype controls are shown. CD14 and CD45 positivity clearly support the monocytic nature of the cells we have obtained. Growing body of evidence suggests that MACs closely resemble to M2-like macrophages which are characterized by anti-inflammatory features as well as to play pro-angiogenic functions . Observations on surface expression of CD34, which may be lost—though not necessarily—during culture, are conflicting. In our study, flow cytometric analysis showed that cells were strongly positive for CD14 and CD45 with weaker expression of the hematopoietic lineage CD34 (which declines through cell culture passages). Our cells also expressed monocyte markers associated with endothelial cell features such as vascular endothelial growth factor receptor-2 ( VEGFR2, also known as KDR) and platelet endothelial cell adhesion molecule-1 (PECAM-1, also known as CD31), suggesting MACs could be considered a sort of educated monocytes . Several groups have reported the coexpression of endothelial markers by these cells. It has been suggested that the detection of endothelial markers might results from contamination with microparticles deriving from other elements in the culture (i.e. platelets) or by passive transfer of platelet microparticles containing CD31 leading to false-positive events by FACS quantification. Testing for CD42 (which, in our hands, was negative) allowed us to exclude contamination with and/or passive transfer of microparticles Lucchesi et al. Cardiovasc Diabetol (2018) 17:78 Page 6 of 13 MACs migration capacity (96.09 ± 4.07%), CD133 (9.27 ± 6.93%) and VEGFR-2 MACs migration was determined in cells from 25 low- (44.15 ± 21.78%) (Fig. 1b). HDL and 23 high-HDL using the 5 µM QCM Chem- otaxis Assay (Millipore, USA), based on the Boyden chamber principle. After 5 days in culture, the cells MACs viability were detached using Trypsin/EDTA and harvested by Viability of the cultured MACs was determined in cells centrifugation; hence, 200,000 cells/well were added to obtained from 48 low-HDL and 40 high-HDL subjects the upper part of a modified Boyden chamber placed in by using the MTT assay. MTT (3-(4,5-Dimethylthiazol- a 24-well culture dish containing EGM-2 and EGM-2 2-yl)-2,5-diphenyltetrazolium bromide) measures mito- enriched with VEGF (50 ng/mL) (Sigma-Aldrich Ltd, chondrial activity in living cells. Briefly, after 5 days of Poole, Dorset, UK) and incubated for 24 h. Cells in the culture MACs were incubated with MTT (Sigma, St. insert were stained and then placed in a well containing Louis, USA) (1 mg/ml) for 3 h at 37 °C, 5%/95% CO / a stain extraction buffer after washing in PBS. Migrating O . Upon incubation, the medium was removed and the cells were counted by colorimetric measurement (optical cells were solubilized in 10% DMSO/90% Isopropanol. density at 540 nm). Then, the amount of the dye released from the cells was quantified by measuring the optical density at 540 nm (reference wavelength: 620 nm) with a multiplate reader ROS production (Multiskan EX, THERMO). The optical density is directly Reactive oxygen species production was evaluated in correlated with the amount of metabolically active cells. cells from 31 low-HDL and 20 high-HDL using ROS- To test the effect of an oxidative stress condition on sensitive fluorescent probe 5-(and-6)-chloromethyl- MACs viability, H O (1 mM, 1 h) was added to the cul- 2 2 2′,7′-dichloro-di-hydro-fluorescein diacetate, acetyl ture medium. ester (CM-H DCFDA) (Invitrogen, Life Technologies Ltd). Briefly, MACs (200,000 cells/well) were incubated with CM-H DCFDA (10 µM/well) for 30 min at 37 °C MACs adhesion to matrix molecules in dark conditions and ROS production was detected as Adhesion capacity was determined in MACs from 21 an increase in fluorescence, by a fluorescence microplate low-HDL and 24 high-HDL. To this purpose MACs reader, at 495 nm excitation and at 527 nm emission. were washed with PBS, and gently detached with 0.25% trypsin/EDTA. After centrifugation and re-suspension, Statistical analysis equal cell numbers (50,000 cells/well) were seeded on Statistical analyses were carried out using the SPSS 13.0 fibronectin coated 96-well microplates, and incubated software (SPSS Inc., Chicago, I, USA) for Mac OS X. for 30 min at 37 °C, 5% C O . The cells were fixed in 4% Data are expressed as median (interquartile range) and/ paraformaldehyde and then incubated with 0.25% crys- or mean ± SD for continuous variables, and number of tal violet for 30 min; therefore, excess dye was removed cases and percentage for categorical variables. Continu- by several washes with PBS and the dye absorbed by ous variables were compared by unpaired Student’s t test adherent cell nuclei was extracted with 33% AcOH. The or by one-way ANOVA (with Scheffe post hoc multiple amount of the dye released from the cells was quantified comparisons) for normally distributed variables and by by measuring the optical density at 540 nm. the Wilcoxon Sum-of-Ranks (Mann–Whitney) U test or the Kruskal–Wallis test for variables with skewed distri- bution. The general linear model (GLM), as an extension MACs senescence of the linear multiple regression for a single dependent Senescence was evaluated in MACs from 21 low-HDL variable (each parameter exploring MACs function), has and 19 high-HDL. Senescent cells were identified using been employed to verify whether HDL levels (the cate- the Senescence Cells Histochemical Staining kit (Sigma- gorical independent factor) still has an effect, beyond the Aldrich Ltd, Poole, Dorset, UK). Briefly, MACs (300,000 effects of covariates for which significant (or marginally cells/well) were washed in PBS, fixed for 7 min at room significant) differences have been observed in low-HDL temperature, washed again and incubated for 16–18 h as compared to high-HDL individuals (gender, age, dia- at 37 °C without CO and with X-gal chromogenic sub- betes duration, BMI, waist circumference, non-HDL cho- strate. After that, cells were washed with PBS, and DMSO lesterol, triacylglycerol, eGFR, presence of hypertension was added to dissolve the stain at 37 °C for 30 min; and smoking habits). Estimated marginal means (ESM) absorbance was measured at 620 nm. To test the effect of have been reported where appropriate. Pearson χ or the an oxidative stress condition on MACs senescence, H O 2 2 Fisher exact probability tests were applied to categorical (1 mM, 1 h) was added to culture medium. Lucchesi et al. Cardiovasc Diabetol (2018) 17:78 Page 7 of 13 variables. Logistic regression analysis with backward categories. MAC studies were conducted in subgroups stepwise variables elimination was used to assess the out of the whole cohort as specified in the “Methods ” independent impact of predictors on each criterion vari- section. able stratified by the median level. p values of 0.05 or less were considered statistically significant. MACs functional properties Results MACs viability Characteristics of the study cohort Viability of MACs obtained from high-HDL patients The main anthropometric and clinical characteristics, was 23% higher than that of cells from low-HDL sub- and the pharmacological treatments of the study cohort jects (111.6 ± 32.7 vs. 90.5 ± 28.6%, p = 0.002). Effect of are shown in Table 1. Compared with individuals with HDL on viability was still significant (ESM 113.0 ± 45.4 low HDL-C those with high HDL-C were more fre- vs. 93.1 ± 42.2%, p = 0.037) beyond the effects of covari - quently females, were older, and had longer diabetes ates and with an independent role for eGFR (p = 0.047). duration and higher total cholesterol levels. Low-HDL To investigate high-HDL contribution to MACs pro- group had higher BMI, waist circumference and preva- tection from oxidative stress, MACs were incubated lence of obesity, higher diastolic BP and prevalence of for 1-h with 1 mM H O . Cell viability decreased in 2 2 hypertension, higher non-HDL cholesterol, triacylglyc- both groups (high-HDL, n. 20: 109.2 ± 31.7% vs. high- erol, uric acid and creatinine levels, with lower eGFR. HDL + H O 74.5 ± 40.8%, p < 0.0001; low-HDL , n. 2 2 Subjects with low-HDL were more frequently current 19: 88.3 ± 25.5% vs. low-HDL + H O 72.3 ± 22.5%, 2 2 smokers and on blood pressure-lowering treatments. p = 0.004; Fig . 2), with a percent decrease in viabil- No differences were found for fasting plasma glucose ity that was significantly higher in MACs from high- levels, HbA1c, systolic BP, LDL cholesterol, treatment HDL than from low-HDL subjects (33.4 ± 24.0% vs . with RAAS (Renin Angiotensin Aldosterone System) 17.1 ± 19.1%; p = 0.025). Viability measured after expo- inhibitors, lipid-lowering agents or glucose-lowering sure to oxidative stress was directly correlated with drugs. Finally, no differences were observed in the prev - viability at baseline in the whole sample (r = 0.655, alence of non-advanced retinopathy, previous major p < 0.0001) as well in high- (r = 0.729, p < 0.0001) and acute cardiovascular events or distribution of UAE low-HDL groups (r = 0.619, p = 0.005) (Fig . 2). 33.4±24.0% 17.1±19.1% ** *** High HDL-ch Low HDL-ch High HDL-ch High HDL-ch Low HDL-ch Low HDL-ch + H O + H O (n.40) (n.48) (n.20) (n.19) 2 2 2 2 (n.20) (n.19) Fig. 2 Ex vivo viability of MACs drawn from type 2 diabetic patients with high HDL-cholesterol levels compared to low HDL-cholesterol group (on the left). MACs viability (on the right) was than evaluated in presence or absence of oxidative stress induced by exposure to H O . The bars 2 2 represent mean ± SD. *p = 0.004; **p = 0.002; ***p < 0.0001 p < 0.05 Optical density, % Lucchesi et al. Cardiovasc Diabetol (2018) 17:78 Page 8 of 13 MACs senescence MACs functional properties by HDL‑cholesterol quartiles Ex vivo senescence assess was comparable in the two To further evaluate the relationship between HDL-C lev- groups (102.7 ± 29.8% vs. 99.2 ± 27.8%, p = 0.703). In els and MACs functional properties, the study population the whole sample, senescence was worsened by expo- was stratified into quartiles (Q) of HDL-C levels yielding sure to H O (96.9 ± 24.5% vs. 89.2 ± 23.4%, p = 0.046); quartiles thresholds of 34, 42 and 71 mg/dl, respectively 2 2 but this difference was not statistically significant (0.88, 1.09 and 1.84 mmol/l). Viability increased from when high- (98.8 ± 23.3% vs. 92.8 ± 22.7%, p = 0.303) Q1 to Q4 (p = 0.001) and was significantly higher in Q4 and low-HDL samples (95.2 ± 26.1% vs. 86.1 ± 24.4%, (124.8 ± 27.1%) than in Q1 (86.9 ± 24.5%, p < 0.001), Q2 p = 0.088) were evaluated separately. Senescence (95.0 ± 33.2%, p = 0.020) and, marginally, than in Q3 in response to oxidative stress was directly corre- (98.8 ± 32.7%, p = 0.055); this remained by comparing lated with senescence at baseline in the whole sam- Q4 and Q1–Q3 (124.8 ± 27.1 vs. 93.3 ± 30.1%, p < 0.0001, ple (r = 0.651, p < 0.0001) as well in high- (r = 0.590, Fig. 3). Effect of HDL quartiles on viability was still sig - p = 0.026) and low-HDL group (r = 0.691, p = 0.003). nificant (p = 0.008) beyond the effects of covariates and with a marginal role for eGFR (p = 0.094). ESM was 128.6 ± 39.3 in Q4 vs. 88.0 ± 36.6 (p < 0.001), 93.2 ± 35.7 (p = 0.006) and 97.1 ± 39.2% (p = 0.004) in Q1–Q3, MACs adhesion capacity to matrix molecules and migration respectively. This difference persisted comparing Q4 and Ex vivo adhesion to fibronectin did not differ in MACs Q1–Q3 (126.7 ± 36.9 vs. 92.4 ± 42.7%, p < 0.001), with from subjects with high-HDL (105.2 ± 32.7%) com- marginal effects for both eGFR (p = 0.092) and diabetes pared to low-HDL (94.1 ± 26.1%, p = 0.223). Also duration (p = 0.090). Consistently, adhesion was higher ex vivo migration capacity was similar in MACs from in Q4 than in Q1–Q3 (115.5 ± 24.1 vs. 95.6 ± 30.4%, subjects with high-HDL (91.3 ± 34.2%) compared to p = 0.044; Fig. 3). This remained marginally significant those with low-HDL (108.7 ± 39.5%, p = 0.111). (ESM 114.4 ± 32.0 vs. 96.0 ± 30.6%, p = 0.057) beyond the effects of covariates. No differences by HDL-C quar - tiles were observed for senescence, migration or ROS ROS production production. Reactive oxygen species production did not differ in MACs from individuals with high-HDL (94.5 ± 30.4%) Independent covariates of MACs functional properties compared to those with low-HDL (103.6 ± 32.0%, HbA1c and HDL-C resulted independent covariates p = 0.316). of MACs viability: indeed, viability increased with the ViabilityAdhesion § § *** ** Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 (n.25) (n.22) (n.22) (n.19) (n.13) (n.11) (n.11) (n.10) Fig. 3 Ex vivo viability and adhesion of MACs drawn from type 2 diabetic patients stratified by HDL-cholesterol level quartiles: Q1 < 34 mg/dl; Q2 34–42 mg/dl; Q3 42–71 mg/dl; Q4 > 71 mg/dl. The bars represent mean ± SD. Viability, on the left (ANOVA one-way p = 0.001): *p = 0.055; § § **p = 0.020; ***p = 0.001; p < 0.0001. Adhesion, on the right: *p = 0.072; p < 0.04 Optical density, % Lucchesi et al. Cardiovasc Diabetol (2018) 17:78 Page 9 of 13 Table 2 Independent correlates of MACs functional properties (logistic regression analysis with stepwise backward variables elimination) Viability Adhesion Senescence OR 95% CI p OR 95% CI p OR 95% CI p BMI (× 1 kg/m ) – – – 0.949 0.908–0.991 0.019 0.878 0.779–0.990 0.033 HbA1c (× 1%) 0.789 0.678–0.919 0.002 – – – 1.683 1.043–2.717 0.033 HDL-cholesterol (× 1 mg/dl) 1.034 1.013–1.055 0.001 1.029 1.005–1.053 0.018 – – – Other variables included in the model, but never Sex, age, waist circumference, diabetes duration, fasting glucose, LDL cholesterol, triacylglycerol, selected as significant independent covariates systolic and diastolic BP, eGFR, albumin-to-creatinine ratio, smoking habits, treatments with BP- lowering and lipid-lowering agents improvement of HbA1c (OR 0.789) as well with the procedures likely enhance vessel formation predomi- increase in HDL-C (OR 1.034) (Table 2). nantly by a paracrine mechanism via a mixture of growth When HDL quartiles were included in the regression factors and cytokines that support angiogenesis, rather instead of HDL-C levels, Q4 was strongly associated than becoming integrated as long-lived endothelial cells with EPCs viability (OR 8.343; 95% CI 2.219–31.369, or directly contributing to re-endothelialization process. p = 0.002). Nevertheless, MACs (“early EPCs”), that are not endothe- Independent predictors of the adhesion capacity were lial nor progenitor cells, albeit cells with pro-angiogenic HDL-C and BMI, i.e. adhesion was directly related to vasoreparative properties , have been suggested as HDL levels (OR 1.029) and inversely with BMI (OR putative biomarkers for cardiovascular disease. Thus, 0.949). When HDL quartiles were included in the regres- ex vivo assessment of MACs function is especially impor- sion instead of HDL-C levels, Q4 was strongly associated tant as these cells are recognized for their role in vascular with MACs adhesion capacity (OR 6.492; 95% CI 1.217– repair in health and disease and have been harnessed as 31.635, p = 0.011). Finally, HbA1c (OR 1.683, directly) therapeutic tools for many ischemic diseases. and BMI (OR 0.878, inversely) emerged as independent Our results can shed some light on the cardiovascular correlates of MACs senescence. protection HDL-cholesterol may exert in type 2 diabetic subjects. Low levels of HDLs are a typical component of Discussion diabetic dyslipidemia . Furthermore, even in those This study shows that in subjects with type 2 diabetes subjects with normal or higher HDL-cholesterol lev- HDL-cholesterol levels can affect MACs functions. In els, modification of the lipoprotein can hamper their particular, ex vivo MACs from patients with high-HDL anti-atherogenic properties [30, 31] including the posi- have higher viability and greater adhesion capacity to tive effect on bioavailability and functional properties matrix molecules compared to MACs from individu- of EPCs [3, 32]. This effect is supported by experimen - als with low-HDL while no difference was apparent with tal evidence. In a mice model i.v. injection of rHDLs respect to senescence, migration capacity, and ROS pro- increased the number of bone-marrow-derived endothe- duction. Finally, we did not observed any concentration- lial cells in the ischemic muscle  and their recruit- related biphasic effects of HDL as previously suggested ment into the murine aortic endothelial layer in response . to an inflammatory insult . In patients with type 2 MACs, originally defined as “early EPCs” or “early diabetes, rHDLs infusion increases plasma HDL anti- outgrowth EPCs” , have been obtained, as recently inflammatory properties, enhances ex vivo cholesterol reviewed , by short-term culturing of PBMCs. efflux capacity , and restores endothelial function as Growth occurs after a few days of adhesion to fibronec - determined by the forearm blood flow response to sero - tin in VEGF-containing medium. This technique yields tonine  as well as the number of circulating EPCs cells with myeloid/hematopoietic characteristics that . The increased number of circulating EPCs as a share features particularly with monocyte/macrophages. result of mobilization of steam/progenitor cells from the MACs, a terminology that clarify both lineage and func- bone marrow to the peripheral blood in response to HDL tion of these cells, are able to promote angiogenesis infusion has been demonstrated in vivo in mouse models in vivo not necessarily by means of an endothelial com- and in humans [11, 33, 35]. On the contrary, the hypoth- mitment, although co-expression of endothelial markers esis of an improved viability and survival suggested by by these cells have been widely reported  and herein our results was only shown using ex vivo assays. This confirmed. Indeed, cells gained by short-term culture last hypothesis is indeed in agreement with the greater Lucchesi et al. Cardiovasc Diabetol (2018) 17:78 Page 10 of 13 ex vivo viability of MACs from type 2 diabetic individuals type 2 diabetic subjects is inversely correlated with with high HDL-C levels (Fig. 2) and is further supported HbA1c, while EPC-bearing clusters inversely correlated by the dose–response relationship between quartiles of with duration of diabetes. HDL-C and MACs viability (r = 0.396, p < 0.001; Fig. 3). We also found an inverse correlation between the In the past, a dual effect of HDLs on EPCs has been MACs adhesion capacity and BMI. This association reported as a function of HDL concentration with is not surprising given the well known effect of body enhanced EPCs tube formation in the presence of low weight on circulating EPCs [42, 43], an effect that levels and a paradoxical increase of senescence and seems to be even greater than the one exerted by blood impaired tube formation at moderate to high concentra- pressure, LDL cholesterol, triacylglycerol, fasting glu- tion . Though we did not see such a clear-cut effect, cose, and smoking . In the study by Heida et al. , it is of interest that the relationship between HDL-C lev- EPCs expanded from obese normo-glycemic subjects els and ex vivo MACs adhesion to matrix molecules is exhibited reduced adhesive, migratory, and angiogenic enhanced only in the highest HDL-C quartile (Fig. 3). capacity, and mice treated with obese-derived EPCs Our results show that HDL-C has no effects on ex vivo showed reduced homing in ischemic hind limbs. Inter- MACs senescence and migration capacity as well on ROS estingly, functional impairment of EPCs was reversible production. The antioxidant capacity of HDLs is mainly after achieving significant weight reduction . In conferred by apolipoproteins and related enzymes such summary, our study explores function of ex vivo MACs as paraoxonases (PON), platelet-activating factor-acetyl and emphasizes the role of body weight and glycemic hydrolase (PAF-AH), glutathione peroxidase (GPx) and control on their senescence. others . In endothelial cells, HDLs can inhibit intra- Several other factors can, obviously, affect MACs func - cellular oxidative stress through inactivation of NADPH tion in diabetes. For instance, the presence, type, and oxidase . In our hands, in MACs from patients with degree of diabetic complications have been reported high HDL-C levels, intracellular ROS generation was to be associated with a whole array of numerical and/ not lower as compared to those from low HDL most or functional impairments of EPCs . Decreased as likely because of impaired anti-inflammatory and anti - well as increased or unchanged EPC number has been oxidant capacity of HDL associated with the lipopro- reported in diabetic patients with severe retinopathy. Of tein modifications occurring in diabetes [36, 37]. MACs note, where increased number of circulating EPCs has are uniquely equipped with intrinsic cellular machinery been reported in patients with diabetic retinopathy, EPC for ROS detoxification. As such, they are more resistant functions such as migration and mobilization or homing to oxidative stress as compared with mature endothelial were often impaired [18, 45]. Moreover, diabetic com- cells [26, 38]. However, impaired EPCs viability and func- plication such as CKD can deeply affect HDLs particle tion as well as increased apoptosis have been observed in number and function. For all these reason, we have paid EPCs under conditions of oxidative stress as generated attention not to include subjects with advanced diabetic by hydrogen peroxide [38–40]. Consistently, exposure retinopathy and those with CKD stages ≥ 3b in this study. of MACs obtained from subjects with type 2 diabetes to Nevertheless, also in these conditions, an independent H O reduced their ex vivo viability (Fig. 2) and acceler- effect of reduced eGFR seems to contribute to impaired 2 2 ated senescence irrespective of HDL-C concentrations MACs viability. Some blood-pressure lowering drugs suggesting those HDL particles may become ineffective (RAAS blockers), lipid-lowering agents (statins) and anti- when challenged by oxidative stress. In support to this hyperglycemic treatments  also may have an effect of hypothesis is the finding that percent reduction viability MACs. However, in our study, all these treatments were in response to H O (Fig. 2) was higher in MACs from evenly distributed in type 2 diabetic subjects with low- 2 2 type 2 diabetes individuals with high HDL-C in spite of and high-HDL cholesterol levels. Moreover, treatments greater viability on basal condition. A similar mecha- did not enter as independent covariates of MACs func- nism may also account for the lack of difference in ex vivo tional properties. EPCs senescence ad migration as previously shown . Because of all this, we are confident that HDL levels A number of mechanisms may support HDL incom- largely mediate the effects we have studied. With respect petence in diabetes including metabolic alterations typ- to this, we have to acknowledge that other features of ically associated with this condition. In line with this these lipoproteins such as their size, composition, and interpretation we found that ex vivo MACs viability is function may have played a role. Epidemiological studies, directly related to HDL-C levels and inversely associ- for instance, suggest that large, buoyant HDL particles ated with HbA1c, but not with diabetes duration. This (i.e. HDL ) may be a better marker of favorable cardio- observation is in keeping with results reported by Tep- vascular outcomes [46, 47] though it has been recently per et al.  showing that proliferation of EPCs from suggested that HDL and HDL cholesterol do not 2 3 Lucchesi et al. Cardiovasc Diabetol (2018) 17:78 Page 11 of 13 necessarily distinguish cardioprotective effects of HDL cell proliferation, migration and reparative properties of subclasses . diabetic cultured myeloid and circulating putative EPCs Some limitations of our study must be taken into con- . Furthermore, in circulating angiogenic cells isolated sideration. First, data on circulating EPCs levels are not from peripheral blood of diabetic subjects it was shown available. This could have provided a complementary that the downregulation of several microRNAs (micro- information on the link between HDL and MACs altera- RNA-155, -126 and -130a) contributes to reduce their tions in type 2 diabetes. Indeed, studies have reported proliferation, promote their senescence and apoptosis, that HDL cholesterol levels correlate with the number of and impair their reparative functions. Apart from micro- circulating CD34+/KDR+ EPCs in subjects with coro- RNAs, other epigenetic mechanisms, including damages nary artery disease , and with the number of CD34+/ and post-translational modifications of DNA, are known CD133+ EPCs in hypercholesterolemic subjects  and to be dysfunctional in diabetes , but, to our knowl- in obese non diabetic women . Furthermore, EPC edge, none of these putative mechanisms has been inves- colony levels were significantly lower in individuals with tigated in relation to the serum concentration of HDL. low HDL in a cohort of patients with cerebrovascular Instead, it is widely recognized that overall HDL in indi- disease including a subgroup of diabetic subjects . In viduals with type 2 diabetes wastes the capacity to sup- a cohort of volunteers with different degrees of glucose press NF-kB-mediated inflammatory response and loses tolerance we have previously reported a significant corre - the ability to stimulate eNOS activation . lation between HDL-cholesterol and circulating CD34+ In particular, it could have been of interest to explore cells but not with CD34+/KDR+ cells . u Th s, to the the role of apolipoprotein A-I (apoA-I) and angiopoie- best of our knowledge, the impact of HDL concentrations tin-like protein 3 (ANGPTL3), a major lipoprotein reg- on circulating EPC levels in diabetes remains a poorly ulator that shows positive correlation with plasma HDL explored field. cholesterol and apoA-I levels. To this regard, it has been Second, ex vivo HDL treatment could have provided recently reported that ANGPLT3 levels are lower in further evidence in support of the role for HDL in MAC female T2DM patients with a weaker association with dysfunction. Literature data show that intravenous infu- HDL components (apoA-I and serum amyloid A) and sion of rHDLs stimulates EPC differentiation and recruit - function (cholesterol efflux) . ANGLPT3 might also ment in rodents [11, 33], exerts beneficial effects on play a role in angiogenesis. Similarly, it could be of value circulating CD34+ cells in patients with recent acute cor- to assess the expression of the scavenger receptor type BI onary syndrome  and increases circulating CD34+/ (SR-BI), a HDL receptor whose deficiency is associated VEGFR2+ cells in patients with type 2 diabetes . Of with impaired HDL function, intracellular cholesterol interest, a long-term trial with Mediterranean diet, that accumulation, increased oxidative stress and regulates reported among other effects a sustained increase in hematopoietic stem/progenitor cells proliferation and HDL cholesterol levels, showed a long-term increase in differentiation . circulating EPCs levels in patients with newly diagnosed type 2 diabetes . Conclusions Another study limitation relates to the lack of in vivo In conclusion, we suggested that MACs derived from assessment of cell function. Previous studies have shown type 2 diabetic individuals with high HDL-C levels show that administration of cultured MACs (“early” EPCs) a relative preservation of some functional properties obtained from diabetic subjects to mice did not promote (mainly viability, to a lesser extent adhesion) as evaluated reendothelialization at site of endothelial injuries nor did ex vivo as compared with cells obtained from subjects restore perfusion in ischemic tissues to the same extent with low HDL-C. Except for viability and adhesion, other of what obtained with cells of healthy controls . How- MACs functions were only marginally or not at all related ever, it is unknown whether MACs derived from diabetic to HDL levels. These properties may be important in pro - subjects with high HDL cholesterol might translate their tecting vascular wall integrity. preserved capability in promoting the formation of func- tional vascular networks in vivo. Abbreviations Finally, we must recognize the lack of mechanistic data AcOH: acetic acid; ACR : albumin-to-creatinine ratio; ALT: alanine aminotrans- on how HDL would impact EPC function in diabetes. ferase; apoA-1: apolipoprotein A-1; AST: aspartate aminotransferase; BMI: body mass index; BP: blood pressure; CKD-EPI: chronic kidney disease epidemiology An increasing number of experiments shows that diabe- collaboration; CM-H2DCFDA: 5-(and-6)-chloromethyl-2′,7′-dichloro-di-hydro- tes impairs the stromal-derived factor-1 (SDF-1)/C-X-C fluorescein diacetate, acetyl ester; CVD: cardiovascular disease; DiI-ac-LDL: chemokine receptor type 4 (CXCR-4) and the nitric oxide 1,1′-dioctadecyl-3,3,3′,3′-tetramethylindocarbocyanine-labeled acetylated low-density lipoprotein; DMSO: dimethyl sulfoxide; EBM-2: endothelial basal (NO)/superoxide anions pathways and the p53/sirtuin1 medium-2; e-CFU: endothelial colony-forming unit; EDTA: ethylenediamine- (SIRT1)/p66Shc axis. All these are major regulators of tetraacetic acid; EGF: endothelial growth factor; eGFR: estimated glomerular Lucchesi et al. Cardiovasc Diabetol (2018) 17:78 Page 12 of 13 filtration rate; EGM-2-MV: microvascular Endothelial Cell Growth Medium‑2; References EMPs: endothelial cell-derived microparticles; EPCs: endothelial progenitor 1. Annema W, von Eckardstein A. High-density lipoproteins. Multifunctional cells; ESM: estimated marginal means; FBS: fetal bovine serum; FGF: fibroblast but vulnerable protections from atherosclerosis. Circ J. 2013;77:2432–48. growth factor; FITC: fluorescein isothiocyanate; FMD: flow-mediated vasodila- 2. Zhang M, Malik AB, Rehman J. Endothelial progenitor cells and vascular tion; GGT : gamma-glutamyltransferase; GLM: general linear model; GPx: repair. Curr Opin Hematol. 2014;21:224–8. glutathione peroxidase; H O : hydrogen peroxide; HbA1c: hemoglobin A1c; 3. Tran-Dinh A, Diallo D, Delbosc S, Varela-Perez LM, Dang QB, Lapergue B, 2 2 HDLs: high-density lipoproteins; HDL-C: HDL-cholesterol; IGF-1: insulin-like et al. HDL and endothelial protection. Br J Pharmacol. 2013;169:493–511. growth factor-1; LDL: low-density lipoprotein; MACs: monocytic angiogenic 4. Rafii S, Butler JM, Ding BS. Angiocrine functions of organ-specific cells; MTT: 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide; NO: endothelial cells. Nature. 2016;529:316–25. nitric oxide; OECs: outgrowth endothelial cells; oxLDL: oxidized LDL; PBMCs: 5. Petoumenos V, Nickenig G, Werner N. High-density lipoprotein exerts peripheral blood mononuclear cells; PBS: phosphate buffered saline; PE: vasculoprotection via endothelial progenitor cells. J Cell Mol Med. phycoerythrin; PAF-AH: platelet-activating factor-acetyl hydrolase; PON-1: par- 2009;13:4623–35. aoxonase-1; RAAS: renin–angiotensin–aldosterone system; rHDL: reconstituted 6. Rossi F, Bertone C, Montanile F, Miglietta F, Lubrano C, Gandini L, et al. HDL; ROS: reactive oxygen species; UAE: urinary albumin excretion; VEGF: HDL cholesterol is a strong determinant of endothelial progenitor cells in vascular endothelial growth factor; VEGFR-2: vascular endothelial growth fac- hypercholesterolemic subjects. Microvasc Res. 2010;80:274–9. tor receptor-2; VPT: vibration perception threshold; X-gal: 5-Bromo-4-chloro-3- 7. Dei Cas A, Spigoni V, Ardigò D, Pedrazzi G, Franzini L, Derlindati E, et al. indolyl ß-d -galactopyranoside. Reduced circulating endothelial progenitor cell number in healthy young adult hyperinsulinemic men. Nutr Metab Cardiovasc Dis. 2011;21:512–7. Authors’ contributions 8. Van Linthout S, Frias M, Singh N, De Geest B. Therapeutic potential DL, SGP, GP, SDP contributed to conception and design of the study; DL, SGP, of HDL in cardioprotection and tissue repair. Handb Exp Pharmacol. VSB contributed to performing experiments and acquisition of data; DL, VSB, 2015;224:527–65. LG, GP, SDP contributed to analysis and interpretation of data; DL, SGP, VSB, LG, 9. Noor R, Shuaib U, Wang CX, Todd K, Ghani U, Schwindt B, et al. High- MG, GP, LP, RM, GP, SDP contributed to drafting the article or revising it criti- density lipoprotein cholesterol regulates endothelial progenitor cally for important intellectual content; DL and SGP contributed equally to this cells by increasing eNOS and preventing apoptosis. Atherosclerosis. study as the first author. All authors read and approved the final manuscript. 2007;192:92–9. 10. Zhang Q, Liu L, Zheng XY. Protective roles of HDL, apoA-I and mimetic Author details peptide on endothelial function: through endothelial cells and endothe- Section of Diabetes and Metabolic Disease, Department of Clinical lial progenitor cells. Int J Cardiol. 2009;133:286–92. and Experimental Medicine, University of Pisa and Azienda Ospedaliero- 11. Tso C, Martinic G, Fan WH, Rogers C, Rye KA, Barter PJ. High-density Universitaria Pisana, Via Paradisa, 2, 56124 Pisa, Italy. Diabetes, Nutrition lipoproteins enhance progenitor-mediated endothelium repair in mice. and Metabolic Diseases, University of Medicine and Pharmacy of Craiova, Arterioscler Thromb Vasc Biol. 2006;26:1144–9. Craiova, Romania. Institute of Agricultural Biology and Biotechnology, 12. Zhang Q, Yin H, Liu P, Zhang H, She M. Essential role of HDL on endothe- National Research Council (CNR), Pisa, Italy. lial progenitor cell proliferation with PI3K/Akt/cyclin D1 as the signal pathway. Exp Biol Med (Maywood). 2010;235:1082–92. Acknowledgements 13. Gordts SC, Van Craeyveld E, Muthuramu I, Singh N, Jacobs F, De Geest We are indebted to the patients attending the Outpatients Diabetic Clinic, and B. Lipid lowering and HDL raising gene transfer increase endothelial to the staff of the “Renzo Navalesi” Diabetes Centre in Pisa particularly to the progenitor cells, enhance myocardial vascularity, and improve diastolic Clinical Laboratory and the Medical Records Unit. function. PLoS ONE. 2012;7:e46849. 14. Huang CY, Lin FY, Shih CM, Au HK, Chang YJ, Nakagami H, et al. Moderate Competing interests to high concentrations of high-density lipoprotein from healthy subjects The authors declare that they have no competing interests. paradoxically impair human endothelial progenitor cells and related angiogenesis by activating Rho-associated kinase pathways. Arterioscler Availability of data and materials Thromb Vasc Biol. 2012;32:2405–17. The datasets used and/or analysed during the current study are available from 15. Vanhoutte PM, Shimokawa H, Feletou M, Tang EH. Endothelial dysfunc- the corresponding author on reasonable request. tion and vascular disease—a 30th anniversary update. Acta Physiol. 2017;219:22–96. Consent for publication 16. Shi Y, Vanhoutte PM. Macro- and microvascular endothelial dysfunction in Not applicable. diabetes. J Diabetes. 2017. https ://doi.org/10.1111/1753-0407.12521 . 17. Rosenson RS, Brewer HB Jr, Ansell BJ, Barter P, Chapman MJ, Heinecke JW, Ethics approval and consent to participate et al. Dysfunctional HDL and atherosclerotic cardiovascular disease. Nat The Ethics Committee of the University of Pisa approved the study protocol Rev Cardiol. 2016;13:48–60. and written informed consent was obtained from all participants before any 18. Fadini GP. A reappraisal of the role of circulating (progenitor) cells in the study procedure. pathobiology of diabetic complications. Diabetologia. 2014;57:4–15. 19. Fadini GP, Ferraro F, Quaini F, Asahara T, Madeddu P. Concise review: dia- Funding betes, the bone marrow niche, and impaired vascular regeneration. Stem This work was supported by a grant from Regione Toscana, Italy, Resolution Cells Transl Med. 2014;3:949–57. 1157 (December 19, 2011), ID Number D55E11002680005. The funder had 20. Sorrentino SA, Besler C, Rohrer L, Meyer M, Heinrich K, Bahlmann FH, no role in study design, data collection and analysis, decision to publish, or et al. Endothelial-vasoprotective effects of high-density lipoprotein are preparation of the manuscript. impaired in patients with type 2 diabetes mellitus but are improved after extended-release niacin therapy. Circulation. 2010;121:110–22. 21. Nieuwdorp M, Vergeer M, Bisoendial RJ, op ‘t Roodt J, Levels H, Birjmohun Publisher’s Note RS, et al. Reconstituted HDL infusion restores endothelial function in Springer Nature remains neutral with regard to jurisdictional claims in pub- patients with type 2 diabetes mellitus. Diabetologia. 2008;51:1081–4. lished maps and institutional affiliations. 22. Stephenson J, Fuller JH, on behalf of the EURODIAB IDDM Complications Study Group. Microvascular and acute complications in IDDM patients: Received: 23 February 2018 Accepted: 21 May 2018 the EURODIAB IDDM Complications Study. Diabetologia. 1994;37:278–85. 23. International Federation of Clinical Chemistry and Laboratory Medicine, IFCC Scientific Division, Mosca A, Goodall I, Hoshino T, Jeppsson JO, John WG, Little RR, et al. Global standardization of glycated hemoglobin measurement: the position of the IFCC Working Group. Clin Chem Lab Med. 2007;45:1077–80. Lucchesi et al. Cardiovasc Diabetol (2018) 17:78 Page 13 of 13 24. Wilkinson CP, Ferris FL 3rd, Klein RE, Lee PP, Agardh CD, Davis M, Global by inhibiting FOXO3a via FOXO3a ubiquitination and degradation. J Cell Diabetic Retinopathy Project Group, et al. Proposed international clinical Physiol. 2015;230:2098–107. diabetic retinopathy and diabetic macular edema disease severity scales. 41. Pan B, Ma Y, Ren H, He Y, Wang Y, Lv X, et al. Diabetic HDL is dysfunctional Ophthalmology. 2003;110:1677–82. in stimulating endothelial cell migration and proliferation due to down 25. Levey AS, Stevens LA, Schmid CH, Zhang YL, Castro AF 3rd, Feldman HI, regulation of SR-BI expression. PLoS ONE. 2012;7:e48530. et al. A new equation to estimate glomerular filtration rate. Ann Intern 42. Müller-Ehmsen J, Braun D, Schneider T, Pfister R, Worm N, Wielckens K, Med. 2009;150:604–12. et al. Decreased number of circulating progenitor cells in obesity: benefi- 26. Lucchesi D, Russo R, Gabriele M, Longo V, Del Prato S, Penno G, et al. Grain cial effects of weight reduction. Eur Heart J. 2008;29:1560–8. and bean lysates improve function of endothelial progenitor cells from 43. Fadini GP, de Kreutzenberg SV, Coracina A, Baesso I, Agostini C, Tiengo A, human peripheral blood: involvement of the endogenous antioxidant et al. Circulating CD34+ cells, metabolic syndrome, and cardiovascular defenses. PLoS ONE. 2014;9:e109298. risk. Eur Heart J. 2006;27:2247–55. 27. Medina RJ, O’Neill CL, O’Doherty TM, Knott H, Guduric-Fuchs J, Gar- 44. Heida NM, Müller JP, Cheng IF, Leifheit-Nestler M, Faustin V, Riggert diner TA, et al. Myeloid angiogenic cells act as alternative M2 mac- J, et al. Eec ff ts of obesity and weight loss on the functional proper - rophages and modulate angiogenesis through interleukin-8. Mol Med. ties of early outgrowth endothelial progenitor cells. J Am Coll Cardiol. 2011;17:1045–55. 2010;55(4):357–67. 28. Fadini GP, Losordo D, Dimmeler S. Critical reevaluation of endothelial 45. Yu CG, Zhang N, Yuan SS, Ma Y, Yang LY, Feng YM, et al. Endothelial progenitor cell phenotypes for therapeutic and diagnostic use. Circ Res. progenitor cells in diabetic microvascular complications: friends or foes? 2012;110:624–37. Stem Cells Int. 2016;2016:1803989. 29. Medina RJ, Barber CL, Sabatier F, Dignat-George F, Melero-Martin JM, 46. de Boer IH, Brunzell JD. HDL in CKD: how good is the “good cholesterol?”. Khosrotehrani K, et al. Endothelial progenitors: a consensus statement on J Am Soc Nephrol. 2014;25:871–4. nomenclature. Stem Cells Transl Med. 2017;6:1316–20. 47. Kontush A. HDL-mediated mechanisms of protection in cardiovascular 30. Vergès B. Pathophysiology of diabetic dyslipidaemia: where are we? disease. Cardiovasc Res. 2014;103:341–9. Diabetologia. 2015;58:886–99. 48. Superko HR, Pendyala L, Williams PT, Momary KM, King SB 3rd, Garrett BC. 31. Rader DJ. Spotlight on HDL biology: new insights in metabolism, func- High-density lipoprotein subclasses and their relationship to cardiovascu- tion, and translation. Cardiovasc Res. 2014;103:337–40. lar disease. J Clin Lipidol. 2012;6:496–523. 32. Tepper OM, Galiano RD, Capla JM, Kalka C, Gagne PJ, Jacobowitz GR, 49. Rossi F, Bertone C, Michelon E, Bianco MJ, Santiemma V. High-density et al. Human endothelial progenitor cells from type II diabetics exhibit lipoprotein cholesterol affects early endothelial progenitor cell number impaired proliferation, adhesion, and incorporation into vascular struc- and endothelial function in obese women. Obesity (Silver Spring). tures. Circulation. 2002;106:2781–6. 2013;21:2356–61. 33. Sumi M, Sata M, Miura S, Rye KA, Toya N, Kanaoka Y, et al. Reconstituted 50. Fadini GP, Pucci L, Vanacore R, Baesso I, Penno G, Balbarini A, et al. Glucose high-density lipoprotein stimulates differentiation of endothelial pro - tolerance is negatively associated with circulating progenitor cell levels. genitor cells and enhances ischemia-induced angiogenesis. Arterioscler Diabetologia. 2007;50:2156–63. Thromb Vasc Biol. 2007;27:813–8. 51. Gebhard C, Rhéaume E, Berry C, Brand G, Kernaleguen AE, Théberge- 34. Patel S, Drew BG, Nakhla S, Duffy SJ, Murphy AJ, Barter PJ, et al. Reconsti- Julien G, et al. Beneficial effects of reconstituted high-density lipoprotein tuted high-density lipoprotein increases plasma high-density lipoprotein (rHDL) on circulating CD34+ cells in patients after an acute coronary anti-inflammatory properties and cholesterol efflux capacity in patients syndrome. PLoS ONE. 2017;12:e0168448. with type 2 diabetes. J Am Coll Cardiol. 2009;53:962–71. 52. Maiorino MI, Bellastella G, Petrizzo M, Gicchino M, Caputo M, Giugliano 35. van Oostrom O, Nieuwdorp M, Westerweel PE, Hoefer IE, Basser R, Stroes D, et al. Eec ff t of a Mediterranean diet on endothelial progenitor cells ES, et al. Reconstituted HDL increases circulating endothelial progeni- and carotid intima-media thickness in type 2 diabetes: follow-up of a tor cells in patients with type 2 diabetes. Arterioscler Thromb Vasc Biol. randomized trial. Eur J Prev Cardiol. 2017;24:399–408. 2007;27:1864–5. 53. Wils J, Favre J, Bellien J. Modulating putative endothelial progenitor cells 36. Morgantini C, Natali A, Boldrini B, Imaizumi S, Navab M, Fogelman AM, for the treatment of endothelial dysfunction and cardiovascular compli- et al. Anti-inflammatory and antioxidant properties of HDLs are impaired cations in diabetes. Pharmacol Ther. 2017;170:98–115. in type 2 diabetes. Diabetes. 2011;60:2617–23. 54. Rajasekar P, O’Neill CL, Eeles L, Stitt AW, Medina RJ. Epigenetic changes in 37. Nobécourt E, Jacqueminet S, Hansel B, Chantepie S, Grimaldi A, Chapman endothelial progenitors as a possible cellular basis for glycemic memory MJ, et al. Defective antioxidative activity of small dense HDL3 particles in in diabetic vascular complications. J Diabetes Res. 2015;2015:436879. type 2 diabetes: relationship to elevated oxidative stress and hypergly- 55. Vaisar T, Couzens E, Hwang A, Russell M, Barlow CE, DeFina LF, et al. Type 2 caemia. Diabetologia. 2005;48:529–38. diabetes is associated with loss of HDL endothelium protective functions. 38. Wang F, Wang YQ, Cao Q, Zhang JJ, Huang LY, Sang TT, et al. Hydrogen PLoS ONE. 2018;13:e0192616. peroxide induced impairment of endothelial progenitor cell viability 56. Zhao D, Yang LY, Wang XH, Yuan SS, Yu CG, Wang ZW, et al. Different is mediated through a FoxO3a dependant mechanism. Microvasc Res. relationship between ANGPTL3 and HDL components in female non- 2013;90:48–54. diabetic subjects and type-2 diabetic patients. Cardiovasc Diabetol. 39. Wang YW, Zhang JH, Yu Y, Yu J, Huang L. Inhibition of store-operated 2016;15:132. calcium entry protects endothelial progenitor cells from H O -induced 57. Gao M, Zhao D, Schouteden S, Sorci-Thomas MG, Van Veldhoven PP, Egg- 2 2 apoptosis. Biomol Ther. 2016;24:371–9. ermont K, et al. Regulation of high-density lipoprotein on hematopoietic 40. Wang YQ, Cao Q, Wang F, Huang LY, Sang TT, Liu F, et al. SIRT1 protects stem/progenitor cells in atherosclerosis requires scavenger receptor type against oxidative stress-induced endothelial progenitor cells apoptosis BI expression. Arterioscler Thromb Vasc Biol. 2014;34:1900–9.
– Springer Journals
Published: Jun 5, 2018