Bone-marrow-derived very small embryonic-like stem cells in patients with critical leg ischaemia: evidence of vasculogenic potential

Endothelium and Vascular Development
Bone-marrow-derived very small embryonic-like stem cells in patients
with critical leg ischaemia: evidence of vasculogenic potential
Coralie L. Guerin1,2,3; Xavier Loyer1,2; José Vilar1,2; Audrey Cras1,3,4; Tristan Mirault1,5; Pascale Gaussem1,3,6;
Jean-Sébastien Silvestre1,2; David M. Smadja1,3,6
1Paris Descartes University, Sorbonne Paris Cité, Paris, France; 2Inserm UMR-S970, PARCC, Paris Research Cardiovascular Research Center, Paris, France; 3Inserm UMR-S1140, Paris,
France; 4AP-HP, Saint Louis Hospital, Cell therapy Department, Paris, France; 5AP-HP, European Georges Pompidou Hospital, Vascular Medicine Department, Paris, France; 6AP-HP,
European Georges Pompidou Hospital, Hematology Department, Paris, France
tern of secretion similar to that of endothelial progenitor cells, as they Very small embryonic-like stem cells (VSELs) are multipotent stem released low levels of VEGF-A and inflammatory cytokines. Note-cells localised in adult bone marrow (BM) that may be mobilised into worthy, VSELs triggered post-ischaemic revascularisation in immuno-peripheral blood (PB) in response to tissue injury. We aimed to quan- deficient mice (p< 0.05 vs PBS treatment), and acquired an endothe- tify VSELs in BM and PB of patients with critical limb ischaemia (CLI) lial phenotype either in vitro when cultured in the presence of VEGF-B and to test their angiogenic potential in vitro as well as their thera- (Cdh-5 gene positive expression), or in vivo in Matrigel implants peutic capacity in mouse model of CLI. We isolated BM VSELs from pa- (human CD31+ staining in neo-vessels from plug sections). In con- tients with CLI and studied their potential to differentiate into vascular clusion, VSELs are a potential new source of therapeutic cells that may
lineages. Flow and imaging cytometry showed that VSEL counts were give rise to cells of the endothelial lineage in humans.
lower in BM (p< 0.001) and higher (p< 0.001) in PB from CLI patients
compared to healthy controls, suggesting that ischaemia may trigger Keywords
VSELs mobilisation in this patient population. Sorted BM-VSELs cul-
Endothelial cells, vasculogenesis, bone marrow, arterial thrombosis, tured in angiogenic media acquired a mesenchymal phenotype stem cells(CD90+, Thy-1 gene positive expression). VSEL-derived cells had a pat- Received: September 9, 2014 Prof. David Smadja Accepted after minor revision: December 7, 2014 European Georges Pompidou Hospital, Hematology Department Epub ahead of print: January 22, 2015 20 rue Leblanc, 75015 Paris, France Tel.: +31 56093933, Fax: +31 56093393 Thromb Haemost 2015; 113: ■■■
are able to recapitulate IH in mice, mainly through their ability to differentiate into endothelial cel s (9, 10). Such cel s have not yet Cel therapy for peripheral artery disease (PAD) is currently based been identified in adults, and the existence of a multipotent adult on the administration of stem and progenitor cell populations progenitor cell (MAPC) has been questioned (11). We suspected thought to contain endothelial progenitor cel s (EPCs) (1, 2). How- that very small embryonic-like stem cel s (VSEL), multipotent ever, EPCs are scarce in circulating blood (3) and have a reduced CD133+ stem cel s found in adult bone marrow (BM), might be a capacity for neovascularisation when isolated from patients with good candidate for cell therapy (12–14). We therefore isolated cardiovascular risk factors (4). Vasculogenic EPCs have been vari- VSELs from patients with critical limb ischaemia and show for the ously quantified as circulating populations expressing surface first time that VSEL-derived cells (VSEL-DCs) expand in proan- markers such as CD34, CD133 and vascular endothelial growth giogenic conditions, can differentiate into vasculogenic cel s, and factor receptor-2 (VEGFR-2), and as "endothelial colony-forming enhance neovascularisation in mice with hindlimb ischaemia.
cel s" (ECFC or "late EPCs") when obtained in culture. However, the ECFC lineage, their CD133 origin, and the relationship be- tween quantification of EPCs by flow cytometry and cel culture Material and methods
procedures is stil debated (5–8).
EPCs appear to be unsuitable for cel therapy, thus alternative Study population and sample collection
sources of progenitor cel s with true vasculogenic potential need to OPTIPEC (ht p://clinicaltrials.gov/ct2/show/NCT00377897) was a be identified. A single postnatal stem cel with endothelial poten- phase I non randomised clinical trial. Patients were eligible for in- tial has so far been identified, namely hemangioma stem cel clusion if they had CLI associated with limited gangrene or a non- (Hem-SC) isolated from infantile hemangiomas (IH). Hem-SCs healing ischaemic ulcer, if they were ineligible for surgical revascu- are CD133+ cel s with a mesenchymal phenotype in culture; they larisation or percutaneous angioplasty, or if such a procedure car- Schattauer 2015 Thrombosis and Haemostasis 113.5/2015 Downloaded from www.thrombosis-online.com on 2015-01-26 ID: 1000471449 IP: 136.165.97.221 Note: Uncorrected proof, epub ahead of print online For personal or educational use only. No other uses without permission. All rights reserved.
Guerin et al. VSEL are vasculogenic cells
ried little chance of success. BM aspirates were collected as de- Cell sorting
scribed elsewhere (15–17). Briefly, BM mononuclear cel s (MNCs) were collected and isolated on Ficoll gradient. Three hours (h) To enrich the CD133+ cel fraction, BM-MNC were magnetical y after cel isolation, 40 injections of 0.75 ml each were made in the label ed with human anti-lineage (Miltenyi Biotec), and cel separ- gastrocnemius muscle of the ischaemic limb. VSEL numbers in ation was achieved by using an AutoMACS device (MiltenyiBio- BM from CLI patients (median age 65 years) were compared to tec) and the Depletes separation program. The negative fraction those isolated from BM samples from control patients (median age was stained as described in the section on MNC flow and imaging 69 years) diagnosed with peripheral thrombocytopenia and free of cytometry phenotyping. Cell sorting was done with a Becton peripheral artery disease or cardiovascular disease. The French Dickinson FACS Aria II SORP device (Becton Dickinson) at a blood bank institute provided venous blood from informed pressure of 20 PSI with a 100-µm nozzle.
healthy donors (agreement with Paris Descartes University, C CPSL UNT N 12/EFS/064). Peripheral venous blood (PB) samples (4 ml) were collected in plastic EDTA-anticoagulant tubes and Cell culture
processed within 4 h. PB was half-diluted in phosphate-buffered Sorted VSELs were seeded on fibronectin-coated 96-wel culture saline (PBS) containing 1 % fetal bovine serum (FBS). BM was plates in complete endothelial cell growth medium-2 (EGM2, one-fourth diluted. Mononuclear cel s (MNC) were isolated by Lonza), 20 % FBS, and were cultured under standard cel culture centrifugation on a 1.077 Ficoll density gradient. After washes, conditions (humidified air-5 % CO2, 37 °C).
MNC were suspended in PBS/1 % FBS to a density of 10×106 cel s Bone-marrow-derived mesenchymal stem cel s (MSCs) were per ml and immediately processed. Frozen BM-MNCs and mobi- obtained from BM-MNCs collected on a Ficoll gradient from lised peripheral blood (mPB) MNCs were stored at –80 °C. healthy volunteer donors. MNCs were plated in six-well plates Samples were quickly thawed in a 37 °C water bath, diluted one- containing 2 ml of endothelial basal medium-2 (Cambrex Bio fifth in PBS/20 % FBS, then washed and suspended in PBS/1 % FBS Science, Walkersvil e, MD, USA) plus 1 % penicil in/streptomycin for immediate processing.
(pen-strep), 10 % FBS and basic fibroblast growth factor (1 ng.mL-1; R&D Systems, Lil e, France). After 48 hours (h), non-ad- MNC phenotyping by flow and imaging cytometry
herent cel s were removed and the medium was changed. Cultures were fed every two or three days until confluence. Expansion after As described in current protocol for VSEL identification (12), the the first passage was performed in EGM2–20 % FBS.
mononuclear cell fraction was counted and stained with anti- Endothelial colony-forming cel s (ECFC) were derived from human lineage-FITC [CD2, CD3, CD14, CD16, CD19, CD24, cord-blood mononuclear cel s as previously described (18). Ex- CD56, CD66b, CD235a (BD Biosciences, San Jose, CA, USA)], pansion was performed in EGM2–20 % FBS.
anti-CD45-PE (BD Biosciences), anti-CD-133/1-APC (Milteny- iBiotec, Bergisch Gladbach, Germany) and DAPI (Sigma Aldrich, St. Louis, MO, USA) to identify dead cel s. Corresponding isotype Cell characterisation
and FMO (fluorescence minus one) controls were performed. We Adherent cultured cel s were washed with PBS, harvested with firstly set the FSC (Forward scat er, representing cel size) and SSC trypsin and EDTA (Invitrogen, Carlsbad, CA, USA) for 5 minutes (Side scatter, representing cell granularity) parameters in logar- (min) at 37 °C, then washed and resuspended with PBS. Subse- ithmic scale. Then, we adjust gating strategy on the FSC parameter quently, cel s were label ed with anti-CD144 phycoerythrin (PE), to include all objects over 2-µm (defined with size calibration anti-CD31 PE-Dyomics 590 and anti-CD90 phycoerythrin (PE)- beads with standard diameters of 1, 2, 4, 6, 10, and 15 µm). Fluor- conjugated monoclonal antibodies for 30 minutes at 4 °C in the escence compensation was achieved with compensation beads, dark. Isotype-matched controls were used. Samples were acquired and with MNC monocolour for DAPI staining. For flow cyto- on an LSRII flow cytometer running FacsDiva software (Becton metry, 2.5 mil ion cel s per sample were acquired on an LSRII flow Dickinson) and were analyzed with Flowjo software (Treestar).
cytometer (Becton Dickinson) running FacsDiva software (Becton Dickinson). Data were analysed with Flowjo software (Treestar, Ashland, OR, USA). Imaging cytometry was performed with a Gene expression analysis
Flowsight device (ImageStream system analysis commercialised by Total RNA was extracted from VSEL-DCs, ECFCs and MCSs by Amnis, Merck Millipore). This ImageStream cytometry analysis using the RNeasy mini kit according to manufacturer's instruc- allows us to better characterise the morphological features of tions (Qiagen, Hilden, Germany). DNA synthesis used the Quanti- VSELs identified. This technique combines classical flow cyto- Tect Reverse Transcription kit (Qiagen). Polymerase chain reac- metry and confocal microscopy analysis with an analysis of these tion amplification was performed on an ABI Prizm 7700 (Applied cel s in suspension. VSELs isolated from PB (controls n=20; CLI Biosystems, Foster City, CA, USA). Cdh5, Von Wil ebrand factor n=30), mobilized PB (controls n=8) and BM mononuclear cel s and Thy1 expression was analysed in TaqMan Gene Expression as- (MNC) (controls n=30; CLI n=11) were identified as lineage- says (Applied Biosystems). The gene expression level was express- CD45– CD133+ live cel s.
ed relative to the glyceraldehyde-3-phosphate dehydrogenase (GAPDH) gene.
Thrombosis and Haemostasis 113.5/2015 Schattauer 2015 Downloaded from www.thrombosis-online.com on 2015-01-26 ID: 1000471449 IP: 136.165.97.221 Note: Uncorrected proof, epub ahead of print online For personal or educational use only. No other uses without permission. All rights reserved.
Guerin et al. VSEL are vasculogenic cells
Figure 1: Gating strategy for sorting of human VSELs derived by are observed by dot plot presenting FSC vs SSC signals. C) Live cel s from (B)
FACS. VSELs were identified within mononuclear cel s (MNCs) according to according to DAPI negative expression are further analysed for haemato-
size criteria, as DAPI- live cel s with a lineage- CD45– CD133+ phenotype by poietic lineages markers expression and Lin– events are isolated (lineage
flow cytometry. FMO and isotypes controls have been done to set up gating negative cel s are defined with FITC mAb for CD2, CD3, CD14, CD16, CD19,
strategy. Percentages on these panels are expressed among previous gated CD24, CD56, CD66b and CD235a). D) The Lin– population from (C) is subse-
population. Percentage showed here is the result from one bone marrow quently analysed based on CD133+ and CD45– antigen expression. Percen-
sample and is representative of al samples. A) Al events ranging above 2 µm tages show the average content of each cel ular subpopulation in total lin-
are considered after comparison with six differently sized bead particles with eage negative cells. E) VSELs are back gated by dot plot presenting FSC vs
standard diameters of 1, 2, 4, 6, 10, and 15 µm. B) Both BM and PB-MNCs SSC. We observe a homogeneous cel population with a size below 10 µm.
and 3 mM inorganic phosphate (for alizarin red staining). The medium was replaced every three days for 14 days.
A total of 30,000 cel s were seeded in 24-wel microplates and cul- Alkaline phosphatase (ALP) activity (▶ Figure 5 C) was deter- tured for 48 h at 37 °C with 1 ml of EGM2–20 % FBS in air-5 % mined by colourimetric conversion of p-nitrophenol from p-nitro- CO2. Supernatants and control media were col ected and cytokines phenylphosphate and compared to values for undifferentiated were quantified by flow cytometry, using compatible plex specific cel s. Cel s were incubated with p-nitrophenol phosphate in alka- for VEGF-A, PDGF-BB, IL-6, IL8 and RANTES (FlowCytomix, line buffer for 30 min at 37 °C. Adding 0.4 M NaOH stopped the Bender MedSystems, Vienna, Austria) according to the manufac- reaction and ALP activity was observed by colouration.
turer's instructions. Data were acquired on an LSRII flow cyto- Mineralisation was detected after cell fixation with ice-cold meter (Becton Dickinson) with 500 recorded events per cytokine 70 % ethanol and Alizarin Red S staining (Amresco, Solon, OH, per sample, and were further analysed with FlowCytomix software USA) to evaluate calcium deposits (▶ Figure 5 D). Cultures were version 3.0 (Bender MedSystems, Vienna, Austria).
rinsed two or three times with PBS to reduce non-specific staining.
Adipogenic cell differentiation
On day 0, 10 000 cel s were seeded in 24-wel microplates and cul- Cel s at 80 % confluence and passage 4–5 were incubated for two tured for 8 h at 37 °C with EGM2–20 % FBS in air-5 % CO2. To weeks in six-wel microplates in adipogenic medium consisting of evaluate growth factor sensitivity, cel layers were washed and in- DMEM with 1 g/l glucose (Gibco, Grand Island, NY, USA), 10 % cubated for 16 h with EBM-2 medium free of serum and growth FBS, 60 µM indomethacin (Sigma), 1 µM dexamethasone, 0.5 mM factors in order to synchronise the cel s. On day 1, after washing 3-isobutyl-1-methylxanthine (IBMX, Sigma), and 1 % penicil in/ with PBS, the medium was replaced by EBM-2/5 % FBS or streptomycin. The medium was replaced every three days for 14 EBM-2/5 % FBS 10 ng.mL-1 VEGF-A or EBM-2 5 % FBS 5 ng.mL- days and fat deposits were observed microscopical y by their typi- 1hFGF or EGM2 20 %FBS, renewed on day 3 and stopped on day cal morphology.
5. Cel s were harvested and counted in a Malassez chamber.
To estimate their proliferative capacity, cel s were cultured with EGM2/20 % FBS in air-5 % CO2 at 37 °C for 14 days. The medium Endothelial cell differentiation
was changed every four days and cel s were counted every other A total of 200,000 cel s per wel were seeded in 24-wel microplates and cultured overnight at 37 °C with EGM2/20 % FBS in air-5 % CO2 and then switched to basal differentiation medium (EBM2, Osteogenic cell differentiation
insulin-transferin-selenium 1X, linoleic acid BSA 1X, dexametha- sone 1 pM, ascorbic acid 2 phosphate 100 µM) plus VEGF-B 10 Eighty %-confluent cultures at passage 4–5 were incubated in six- ng.mL-1. Controls lacked VEGF-B.
well microplates in osteogenic medium consisting of Dulbecco's modified Eagle's medium (DMEM, Invitrogen) with 4.5 g.L-1 glu- cose, 10 % FBS, 10–7 M dexamethasone, 50 µg.mL-1 ascorbic acid, Schattauer 2015 Thrombosis and Haemostasis 113.5/2015 Downloaded from www.thrombosis-online.com on 2015-01-26 ID: 1000471449 IP: 136.165.97.221 Note: Uncorrected proof, epub ahead of print online For personal or educational use only. No other uses without permission. All rights reserved.
Guerin et al. VSEL are vasculogenic cells
Figure 2: Gating strategy for human VSELs by imaging cytometry. image intensity profile (brightfield) above 60. A) Expression of lineage +
The VSEL phenotype was confirmed by imaging cytometry analysed by the CD45+ cells. B) Expression of lineage – CD45+ cells. C) Expression of lin-
ImageStream system. We observed several expression patterns of MNCs eage-CD45+CD133+ cells. D) Expression of lineage-CD45-CD133+ cells.
selected on gradient Root mean square (RMS) of the rate of change of the These are VSELs confirmed at the single-cel level.
Hindlimb ischaemia model
X-ray transducer. The images were assembled to obtain a complete view of the hind limbs. The number of pixels occupied by vessel Nude mice underwent permanent right femoral artery ligation was measured in the area of quantification by using Primedangio under isoflurane anesthesia on day 0. On day 1, 100,000 cel s sus- software (Trophy System, Paris, France). The area of quantification pended in PBS were injected into both the tibia anterior and ga- was limited by placement of the ligature on the femoral artery, the strocnemius muscles of anesthetised mice. Laser Doppler perfu- knee, the edge of the femur, and the external limit of the leg. The sion imaging and microangiography were performed on day 10. results were expressed as the ratio of ischaemic to non-ischaemic For Doppler imaging, mice were placed on a heating plate at 37 °C leg vessel number.
to minimise temperature variations, and foot perfusion was measured with a Moor LDI. Results were expressed as the ratio of ischaemic to non-ischaemic foot perfusion. The mice were then In vivo Matrigel plug assay
anesthetised with pentobarbital, and longitudinal laparotomy was Matrigel plugs (0.3 ml) were prepared on ice by mixing Matrigel performed to introduce a polyethylene catheter into the abdomi- (BD Matrigel Matrix, BD Biosciences, Le Pont de Clais, France) nal aorta in order to inject contrast medium (barium sulfate, 1 with PBS or with 3×106 cel s suspended in PBS, and were then im- g.mL-1). Angiography of the hind limbs was then performed, and planted in the back of nude mice. After 14 days the plugs were images (2 per animal) were acquired with a high-definition digital Figure 3: Critical limb ischemia triggers
VSEL mobilisation from bone marrow to
peripheral blood. Normal VSEL numbers
among lineage-negative MNCs was determined
in peripheral blood (PB) and bone marrow (BM)
from healthy controls. Significantly higher
numbers were observed in mobilised blood (mB)
from control cytapheresis and in PB from CLI pa-
tients. Numbers of VSELs were lower in BM from
CLI patients than controls. This imbalance in
VSEL numbers in CLI patients, with an increase
in PB and a decrease in BM, suggests that CLI
triggers VSEL mobilisation from BM to PB. Con-
trol PB n=20, mB n=8, BM n=35; CLI PB n=30,
BM n=11. Mann-Whitney test, *** p< 0.001.
Thrombosis and Haemostasis 113.5/2015 Schattauer 2015 Downloaded from www.thrombosis-online.com on 2015-01-26 ID: 1000471449 IP: 136.165.97.221 Note: Uncorrected proof, epub ahead of print online For personal or educational use only. No other uses without permission. All rights reserved.
Guerin et al. VSEL are vasculogenic cells
careful y removed, fixed, and embedded in paraffin for histologic notype of VSELs, ImageStream imaging analysis allowed us to show the decreased size of VSELs in contrast to CD45+/lineage- Blood vessel infiltration was quantified by hematoxylin and negative or positive cel s (▶ Figure 2). Among lineage-negative
eosin (H&E) staining. Three sections of each plug were analysed. MNCs, VSELs were detectable in healthy volunteers' peripheral Paraffin sections were stained with a human-CD31-specific mouse blood (0.005 % ± 0.0004) and bone marrow (0.31 % ± 0.05) monoclonal antibody (1:100, Becton Dickinson) and revealed with (60-fold higher in BM, p< 0.001). VSEL counts were significantly goat anti-mouse antibody Alexa-488 (1:400, Invitrogen) to detect higher in samples obtained by cytapheresis of mobilized peripheral microvessels, while perivascular cells were stained with antibody blood (PB) (0.28 % ± 0.08; p< 0.001 vs control PB). VSEL counts αSMA-Cy3 (1:1,000, Sigma). The sections were mounted with were significantly higher in PB of CLI patients than controls Fluoprep (Biomerieux, Marcy l'Etoile, France) supplemented with (0.016 % ± 0.002; p< 0.001), and significantly lower in BM of CLI 2 µg.ml-1 DAPI. Fluorescence was examined with a Leica patients than controls (0.06 % ± 0.01; p< 0.001) (▶ Figure 3). These DMI-6000 microscope (Leica, Wetzlar, Germany).
results suggest that VSEL are mobilised from BM to PB in CLI pa- tients, as previously described in patients with acute myocardial infarction (19).
Data were analyzed using the non-parametric Mann-Whitney VSEL-DCs acquire a mesenchymal phenotype in
t-test or an ANOVA fol owed by Fisher's protected least-significant proangiogenic conditions
difference test. Differences were considered significant at p< 0.05. All statistical analyses were performed with GraphPad Prism 5 BM VSELs from CLI patients were sorted and cultured in com- software (GraphPad Software Inc., San Diego, CA, USA) and Stat plete endothelial growth medium (▶ Figure 4 A). Human ECFCs View software package (SAS, Cary, NC, USA).
and MSCs cultured in identical conditions were used as controls. Flow cytometric analysis of endothelial and mesenchymal cel sur- face markers showed that VSEL-DCs acquired a mesenchymal phenotype with negative CD144 and CD31 expression (▶ Figure Bone marrow VSELs are mobilised during critical limb 4 B) and positive CD90 expression (▶ Figure 4 C), as well as a
negative Cdh-5 and Von Willebrand factor gene expression (▶ Figure 4 D) and a positive Thy-1 gene expression (▶ Figure VSELs were identified by flow (▶ Figure 1) and imaging (▶ Figure 4 E). VSEL-DCs were also able to differentiate into adipocytes and 2) cytometry as live cel s of appropriate size and with a lineage- osteocytes (▶ Figure 5) in wel -established conditions of mesen- negative, CD45-, CD133+ phenotype. More than confirming phe- chymal differentiation (20). Adipocytes differentiation was de- Figure 4: VSEL-DCs acquire a mesenchymal phenotype in angiogenic (mesenchymal marker) in angiogenic culture conditions. D) CDH5 (VE-
conditions. A) Phase-contrast image of VSEL-DCs cultured in angiogenic cadherin) and Von Willebrand factor gene expression in cultured ECFCs,
conditions. B) VSEL-DCs were negative for CD144 and CD31 (endothelial MSCs and VSEL-DCs. E) Thy1 gene expression in cultured ECFCs, MSCs and
maker) in angiogenic culture conditions. C) VSEL-DCs were positive for CD90 VSEL-DCs. Data are means ± SEM. n=3 for each condition.
Schattauer 2015 Thrombosis and Haemostasis 113.5/2015 Downloaded from www.thrombosis-online.com on 2015-01-26 ID: 1000471449 IP: 136.165.97.221 Note: Uncorrected proof, epub ahead of print online For personal or educational use only. No other uses without permission. All rights reserved.
Guerin et al. VSEL are vasculogenic cells
tected after 14 days in adipogenic medium (in particular adipo- VSEL-DCs are vasculogenic cells
genic inducer 3-isobutyl-1-methylxanthine, ▶ Figure 5 A and B). Osteogenic differentiation ability was firstly revealed in ▶ Figure We then examined the therapeutic potential of VSEL-DCs in a 5 C by ALP activity, determined by colourimetric conversion of preclinical model of CLI. Compared to PBS (control), laser p-nitrophenol from p-nitrophenylphosphate, then explored in Doppler measurement (▶ Figure 7 A) and angiography (▶ Figure ▶ Figure 5 D by studying mineralisation, detected by Alizarin Red 7 B) showed a significant positive effect of VSEL-DCs on blood S staining to evaluate calcium deposits. This capacity for differ- reperfusion in the mouse hindlimb ischaemia (▶ Figure 7 C), simi- entiation indicates that VSEL-DCs are multipotent cel s. We then lar to that of ECFCs and MSCs. We then examined whether VSEL- compared the growth capacity of VSEL-DCs to that of MSCs and DCs could form vessels in immunodeficient nu/nu mice, using ECFCs. VSEL-DCs showed significantly lower proliferative poten- Matrigel implants as previously described (21, 22). VSEL-DCs tial than MSCs and ECFCs (▶ Figure 6 A). The proliferative re- contained in Matrigel (3×106 cel s/200 µl of Matrigel/animal) were sponse to serum, basic bFGF and VEGF-A was also determined implanted subcutaneously in eight-week-old mice. Fourteen days after 16 h of serum and growth factor starvation (▶ Figure 6 B). after implantation, hematoxylin and eosin staining of Matrigel sec- VSEL-DCs proliferated in response to serum and bFGF. VSEL- tions showed numerous vascular channels filled with red blood DCs and MSCs did not proliferate in response to VEGF-A, while cel s, which were more abundant than in ECFC- and MSC-treated ECFCs proliferated as expected. Because MSCs mainly promote Matrigel (▶ Figure 8 A). Fluorescence staining showed that VSEL- vessel growth through their paracrine activity (16), we also as- DCs were incorporated into vessel walls and expressed human sessed growth factor and cytokine secretion by VSEL-DCs. Sur- CD31, contrary to MSCs (▶ Figure 8 B). This ability to differenti- prisingly, their pat ern of cytokine secretion is comparable to that ate into endothelial cel s was confirmed in vitro. Indeed, VSEL- of ECFCs, and they released less VEGF-A (▶ Figure 6 C) and in- DCs cultured at high density (2×105 cel s/cm2) for eight days in flammatory cytokines than MSCs (▶ Figure 6 D).
endothelial differentiation medium (serum-free medium, plus VEGF-B 10 ng.mL-1, as previously described for Hem-SC (10)) ex- Figure 5: VSEL-DC are multipotent in vitro. A) Phase-contrast image of Alkaline phosphatase (ALP) activity (C) was determined by colourimetric con-
untreated VSEL-DCs. B) Adipogenic cell differentiation and fat deposition version of p-nitrophenol from p-nitrophenylphosphate. Mineralisation was
revealed by typical morphology. C, D) Osteogenic cel differentiation revealed detected after cell fixation with ice-cold 70 % ethanol and Alizarin Red S
by alkaline phosphatase activity and mineralisation. n=3 for each condition staining to evaluate calcium deposits (D).
and each experiment has been done with BM-MSCs as a positive control.
Thrombosis and Haemostasis 113.5/2015 Schattauer 2015 Downloaded from www.thrombosis-online.com on 2015-01-26 ID: 1000471449 IP: 136.165.97.221 Note: Uncorrected proof, epub ahead of print online For personal or educational use only. No other uses without permission. All rights reserved.
Guerin et al. VSEL are vasculogenic cells
pressed VE-cadherin. In contrast, VE-cadherin expression was not mesenchymal phenotype. VSEL-derived cel s triggered post-is- detected in the absence of VEGF-B (▶ Figure 8 C).
chaemic revascularisation in immunodeficient mice, and were able to differentiate into cel s with endothelial phenotype in Matrigel implants in vivo or when cultured with VEGF-B in vitro. Alto- gether, our results demonstrate that VSELs are precursors of en- dothelial lineage cel s and thus may represent a potential new In the present study, we unravel for the first time the presence of source of therapeutic cel s.
VSELs in bone marrow and peripheral blood of patients with CLI, EPCs have been proposed for autologous pro-angiogenic ther- these cel s being mobilised from BM to blood in this ischaemic apy and display the ability to migrate to ischaemic tissue and to context. Sorted VSELs cultured in angiogenic media acquired a contribute to neovascularisation in response to tissue ischaemia Figure 6: VSEL-DCs show a pattern of cytokine secretion resembled similarity is supported by a non-significant trend but nevertheless interesting
that of ECFCs. A) VSEL-DCs showed a low proliferative capacity. B) VSEL-
PDGF-BB secretion comparable to that of ECFC. ns, non significant, *** DCs have the same profile as MSCs and are sensitive to bFGF. Data are fold p< 0.001. D) As ECFCs, VSEL-DCs do not secrete inflammatory cytokines IL-6, changes vs control conditions (EBM 5 %) ± SEM. Mann-Whitney t-test, * IL-8 and RANTES compared to MSCs (p< 0.001 vs ECFCs or VSEL-DCs). ns, p< 0.05, ** p< 0.01 vs EBM 5 %. C) VSEL-DCs secreted the same pattern of non significant, *** p< 0.001. n=3 independent experiments, with each con-growth factors as ECFCs with, for both cell types, no significant level of dition in triplicate.
VEGF-A produced compare to MSCs (p< 0.001 vs ECFCs or VSEL-DCs). This Schattauer 2015 Thrombosis and Haemostasis 113.5/2015 Downloaded from www.thrombosis-online.com on 2015-01-26 ID: 1000471449 IP: 136.165.97.221 Note: Uncorrected proof, epub ahead of print online For personal or educational use only. No other uses without permission. All rights reserved.
Guerin et al. VSEL are vasculogenic cells
(2). However, the specific population or sub-population of EPCs set ing as a cel therapy strategy (30–32). Thus, a major step-for- responsible for the resultant pro-angiogenic effect and the subse- ward in the development of second generation of cel -based strat- quent clinical benefit remains unidentified. Alternatively, the dis- egy should be the identification of stem/progenitor that may un- covery of BM-derived adult progenitors with postnatal angiogenic deniably incorporate newly formed vessels in adult ischaemic mi- potential leads to the development of a cel therapy approach to lieu. Several surface antigens have been used in an at empt to en- treat PAD and particularly its more severe form, CLI. The Thera- rich cell therapy product with EPCs and ECFCs, Hence, isolated peutic Angiogenesis by Cel Transplantation (TACT) study was the hematopoietic CD34+ cel s gave rise to positive endothelial first trial to demonstrate a significant improvement fol owing in- marker expressing cel s in vitro and in vivo (33). However, the epi- tramuscular injection of autologous BM concentrated mononu- genetic status of CD34+ and CD34+KDR+ cel s displays a high clear cel s (BM-MNC) (23). In the French Optimisation of Pro- level of DNA methylation of the endothelial nitric oxide synthase genitor Endothelial Cel s in the Treatment of Critical Leg Ischae- promoter as well as histone modifications of several endothelial mia (OPTIPEC) trial, the development of a proliferative angio- marker genes, suggesting that these cel s are not predisposed to ac- genic process was observed in the toe or trans-metatarsal ampu- quire an endothelial cel fate without specific adequate reprogram- tation specimens of BM-MNC-treated patients (15, 18). BM-MNC ming program (34). As CD34 antigen is stil expressed by mature and circulating MNC contain a complex assortment of angiogenic endothelial cel s and because ECFCs express CD133 mRNA dur- cel s, in which the active cel ular components remain to be deter- ing the first weeks of culture (35, 36), other studies have used mined. Among them, ECFC (3), also known as ‘late' EPC, develop CD133 to select the bona-fide EPCs (37). CD133 antigen positive after 2–3 weeks of culture and have the characteristics of precur- cel s at the origin of endothelial lineage have also been criticised, as sors commit ed to an endothelial lineage. They are promising can- two groups (5, 6) were unable to induce the differentiation of didate cel s as they have the capacity to incorporate into neovessels CD133+ cel s into endothelial cel s. Conversely, several convincing (3). ECFCs have never been isolated from cultured BM, whereas papers have shown that peripheral blood–, bone marrow–, or um- they have been obtained from the blood of patients with CAD bilical cord blood–derived CD34+ or CD133+ cel s are enriched in (24), acute myocardial infarction (MI) (25), pulmonary hyperten- cel s of endothelial lineage, express endothelial marker genes, and sion (26, 27) and pulmonary fibrosis (28), but their therapeutic form endothelial structures in vitro and in vivo (38, 39). Neverthe- potential and their implication as a biomarker have yet to be ful y less, conflicting results also arise from additional studies showing explored and understood. In PAD patients, we and others have that the angiogenic effect of PB-, BM- or cord blood-derived found a decreased number of circulating EPC, and especially CD133+ cel s mainly relies on the paracrine secretion of numerous ECFCs (17, 29). In addition, the low ECFC numbers obtained growth factors and vasoactive molecules (40).
from the blood of CLI patient and their poor clonogenic and pro- In infantile haemangiomas (IH), CD133+ cel s have been liferative properties dampen their potential usefulness in clinical shown to give rise to stem cel s with endothelial potential, namely Figure 7: VSEL-DCs promote neovascularisation. A) Laser Doppler DCs efficiently restored blood perfusion and promoted neovascularisation.
measurements lead to evaluate blood perfusion. B) Microangiography shows n=3 independent experiments, with each condition in triplicate. Mann-Whit-
neovascularisation. C) Ten days after the onset of hindlimb ischaemia, VSEL-
ney t-test, * p< 0.05, ** p< 0.01 vs PBS.
Thrombosis and Haemostasis 113.5/2015 Schattauer 2015 Downloaded from www.thrombosis-online.com on 2015-01-26 ID: 1000471449 IP: 136.165.97.221 Note: Uncorrected proof, epub ahead of print online For personal or educational use only. No other uses without permission. All rights reserved.
Guerin et al. VSEL are vasculogenic cells
haemangioma stem cel s (Hem-SC). Hem-SCs isolated from ischaemia model, we previously described that human MSCs were CD133+ cel s have a mesenchymal phenotype in culture and the not able to form chimeric vessels containing endothelial cel s from ability to differentiate into endothelial cel s (9, 10). No equivalent both human and murine origin (16). We confirm here this result mesenchymal subtypes have been identified in adults, since there in a matrigel implant. In vivo, MSCs did not induce vessel is no convincing proof for an endothelial differentiation of mesen- formation of human origin and differentiated cultured MSCs did chymal cel s. hVSEL cel s are a resident population of multipotent not express VE-cadherin after VEGF-B treatment, in contrast to stem cel s in the bone marrow that express CD133+ cel s and VSEL-DCs or Hem-SCs (10). Thus, VSEL-DCs, despite the same pluripotency markers including Oct-4 and Nanog. These cel s mesenchymal phenotype than MSC, are BM-mesenchymal de- have been described to give rise to cel s from al the three germ lin- rived cel s that have robust vasculogenic potential with a possibil- eages (ectoderm, mesoderm and endoderm) (41, 42) and a cardio- ity to give rise to endothelial cel s. Another difference between vascular cel differentiation has been demonstrated in the mouse MSCs and VSEL-DCs is their secretion profile. MSCs have been (43). The results of the present study establish that hVSEL cel s can shown to support angiogenesis via release of angiogenic factors, in be mobilised from BM to PB in CLI patients. Human BM VSEL particular VEGF-A (44, 45). Our present data show that VSEL- isolation and differentiation in vitro under angiogenic conditions DCs, despite a mesenchymal phenotype and negative expression of al owed us to obtain so-cal ed VSEL-derived cel s (DC), able to in- endothelial markers CD144, CD31 or von wil ebrand factor, have a duce revascularisation in hind limb ischaemia model and matrigel pattern of secretion similar to that of EPCs and release low implant. In Matrigel implanted with VSEL-DC alone, we also amounts of VEGF-A and inflammatory cytokines.
showed that a subsequent number of endothelial cel s results from It has been proposed that VSELs represent a quiescent popu- the differentiation of human VSEL-DC and not from stem/pro- lation of cel s that reside within the BM (14), and that after their genitor cel s of the host animal. VSEL-DCs have the same pheno- mobilisation into peripheral blood, they are able to participate in type in vitro than MSCs characterised by a CD90 expression and the turnover of other tissue-specific stem cel s, impacting then tis- ability to differentiate into different germ lineages. In the hindlimb sue organ regeneration (46, 47). Thus, while VSELs are present at Figure 8: VSEL-DCs are able to differentiate into endothelial cells. A) gin, suggesting that VSEL-DCs are able to differentiate into endothelial cel s
Hematoxylin and eosin staining of Matrigel plug sections seeded with 3×106 in situ contrary to human bone marrow MSCs. C) Endothelial differentiation
VSEL-DCs, showing a larger number of functional vessels than were obtained of VSEL-DCs was confirmed in vitro by culture with VEGF-B, revealing specific
with MSCs and ECFCs. Data are means ± SEM. B) Neovessels in Matrigel Cdh5 (endothelial marker) expression. Each figure shows the results of three
plugs seeded with VSEL-DCs, showing CD31+ endothelial cel s of human ori-
independent experiments, with each condition tested at least in triplicate.
Schattauer 2015 Thrombosis and Haemostasis 113.5/2015 Downloaded from www.thrombosis-online.com on 2015-01-26 ID: 1000471449 IP: 136.165.97.221 Note: Uncorrected proof, epub ahead of print online For personal or educational use only. No other uses without permission. All rights reserved.
Guerin et al. VSEL are vasculogenic cells
very low level in the blood during physiological set ing, their abil- ate into endothelial cel s may provide an avenue for deciphering ity to mobilise into blood after the onset of ischaemia is particu- the ontogeny of the endothelial lineage.
larly attractive from the standpoint of developing a therapeutic product. In this line, VSELs are mobilised into peripheral blood Acknowledgements
fol owing acute myocardial infarction (19) and stroke (48). Simi- We thank Livine Duban for her excellent management of the larly, VSEL circulating number is markedly increased in CLI pa- Flowsight image cytometer. This work was supported by grants tients, supporting the concept that these cel s may participate to from Région Ile de France-CORDDIM (Domaine d'intérêt majeur the vascular repair after an ischaemic insult. As a prerequisite to Cardiovasculaire Obésité Rein Diabète) and the Conny-Maeva clinical trials, it is also essential to determine the hVSEL expansion Charitable Foundation.
potential. In our culture conditions, the ability of VSEL-DCs to proliferate in vitro was limited. Hence, further work is required to Author contributions
stimulate the proliferative capacity of VSELs without affecting C. L. G. wrote the manuscript and performed al experiments with their potential for differentiation as wel as to elucidate the exact help from X. L., J. V. and A. C.; T. M. provided samples; P. G. and molecular mechanism of their endothelial differentiation. VSEL- J. S.S discussed the results, provided ideas and critical y read the DCs described here are the first adult-derived stem cel s with a ca- manuscript; D. M. S. conceived the study and wrote the manu- pacity for in vitro differentiation equal to that of stem cel s isolated script.
from IH. Indeed, VSEL-DCs isolated from CD133+ cel s recapitu- late endothelial formation observed in Hem-SCs, however with a Conflicts of interest
lower proliferative potential. Next step wil be to elucidate differ- None declared.
ence between these cel types, in particular to address if the prolif- eration property results from pathophysiology of infantile haem- angioma tumour, cel immaturity or patient's age.
In conclusion, we show here that VSELs have angiogenic prop- 1. Fadini GP, Losordo D, Dimmeler S. Critical reevaluation of endothelial progeni- erties in vivo and are able to differentiate into endothelial lineage. tor cell phenotypes for therapeutic and diagnostic use. Circ Res 2012; 110: VSELs acquire a mesenchymal cel phenotype in vitro, with a re- stricted paracrine potential. However, they play a role in neovascu- 2. Silvestre JS, Smadja DM, Levy BI. Postischaemic revascularisation: from cel ular larisation by improving microvessel density and tissue perfusion in and molecular mechanisms to clinical applications. Physiol Rev 2013; 93: a mouse model. The unparal eled capacity of VSELs to differenti- 3. Yoder MC, Mead LE, Prater D, et al. Redefining endothelial progenitor cel s via clonal analysis and hematopoietic stem/progenitor cell principals. Blood 2007; 109: 1801–1809.
4. Heeschen C, Lehmann R, Honold J, et al. Profoundly reduced neovasculari- sation capacity of bone marrow mononuclear cel s derived from patients with What is known about this topic?
chronic ischaemic heart disease. Circulation 2004; 109: 1615–1622.
• Bone marrow (BM)-derived stem and progenitor cells have been 5. Case J, Mead LE, Bessler WK, et al. Human CD34+AC133+VEGFR-2+ cels are
not endothelial progenitor cel s but distinct, primitive hematopoietic progeni- identified as a potential new therapeutic option to induce thera- tors. Exp Hematol 2007; 35: 1109–1118.
peutic angiogenesis in critical limb ischaemia.
6. Timmermans F, Van Hauwermeiren F, De Smedt M, et al. Endothelial out- • Very small embryonic-like stem cells (VSELs) are multipotent stem
growth cel s are not derived from CD133+ cel s or CD45+ hematopoietic pre- cursors. Arterioscler Thromb Vasc Biol 2007; 27: 1572–1579.
cells localised in adult BM.
7. Mauge L, Sabatier F, Boutouyrie P, et al. Forearm ischaemia decreases endothe- Ontogeny of endothelial lineage in adult is not fully understood lial colony-forming cell angiogenic potential. Cytotherapy 2014; 16: 213–224.
and a postnatal adult stem cell with endothelial commitment still 8. George J, Shmilovich H, Deutsch V, et al. Comparative analysis of methods for needs to be explored.
assessment of circulating endothelial progenitor cel s. Tissue Eng 2006; 12: 9. Khan ZA, Boscolo E, Picard A, et al. Multipotential stem cel s recapitulate What does this paper add?
human infantile hemangioma in immunodeficient mice. J Clin Invest 2008; 118: • We described a VSELs mobilisation from BM to peripheral blood
in this critical limb ischaemia.
10. Boscolo E, Mulliken JB, Bischoff J. VEGFR-1 mediates endothelial differenti- ation and formation of blood vessels in a murine model of infantile hemangio- Sorted BM-VSELs cultured in angiogenic media acquired a mesen- ma. Am J Pathol 2011; 179: 2266–2277.
chymal with a pattern of secretion similar to that of endothelial 11. Jiang Y, Vaessen B, Lenvik T, et al. Multipotent progenitor cel s can be isolated progenitor cells.
from postnatal murine bone marrow, muscle, and brain. Exp Hematol 2002; 30: These VSEL derived cells triggered post-ischaemic revasculari- 12. Zuba-Surma EK, Ratajczak MZ. Overview of very small embryonic-like stem sation in immunodeficient patients and acquired an endothelial cel s (VSELs) and methodology of their identification and isolation by flow cy- phenotype both in vitro and in vivo. This endothelial differenti- tometric methods. Curr Protoc Cytom 2010; Chapter 9: Unit9 29.
13. Ratajczak MZ, Zuba-Surma EK, Shin DM, et al. Very small embryonic-like ation of adult stem cells provide first evidence that VSEL are a po- (VSEL) stem cel s in adult organs and their potential role in rejuvenation of tis- tential new source of therapeutic cells that may give rise to cells sues and longevity. Exp Gerontol 2008; 43: 1009–1017.
of the endothelial lineage in humans, in particular in critical limb ischaemia patients.
Thrombosis and Haemostasis 113.5/2015 Schattauer 2015 Downloaded from www.thrombosis-online.com on 2015-01-26 ID: 1000471449 IP: 136.165.97.221 Note: Uncorrected proof, epub ahead of print online For personal or educational use only. No other uses without permission. All rights reserved.
Guerin et al. VSEL are vasculogenic cells
14. Ratajczak MZ, Zuba-Surma EK, Machalinski B, et al. Very small embryonic-like 32. Smadja DM, Mauge L, Susen S, et al. Blood outgrowth endothelial cel s from (VSEL) stem cel s: purification from adult organs, characterisation, and biologi- cord blood and peripheral blood: angiogenesis-related characteristics in vitro: a cal significance. Stem Cell Rev 2008; 4: 89–99.
rebuttal. J Thromb Haemost 2009; 7: 504–506; author reply 506–508.
15. Duong-Van-Huyen JP, Smadja DM, Bruneval P, et al. Bone marrow-derived 33. Asahara T, Murohara T, Sullivan A, et al. Isolation of putative progenitor en- mononuclear cell therapy induces distal angiogenesis after local injection in dothelial cel s for angiogenesis. Science 1997; 275: 964–967.
critical leg ischaemia. Mod Pathol 2008; 21: 837–846.
34. Ohtani K, Vlachojannis GJ, Koyanagi M, et al. Epigenetic regulation of endothe- 16. Smadja DM, d'Audigier C, Guerin CL, et al. Angiogenic potential of BM MSCs lial lineage committed genes in pro-angiogenic hematopoietic and endothelial derived from patients with critical leg ischaemia. Bone Marrow Transplant progenitor cel s. Circ Res 2011; 109: 1219–1229.
2012; 47: 997–1000.
35. Bompais H, Chagraoui J, Canron X, et al. Human endothelial cel s derived from 17. Smadja DM, Duong-van-Huyen JP, Dal Cortivo L, et al. Early endothelial pro- circulating progenitors display specific functional properties compared with genitor cel s in bone marrow are a biomarker of cell therapy success in patients mature vessel wall endothelial cel s. Blood 2004; 103: 2577–2584.
with critical limb ischaemia. Cytotherapy 2012; 14: 232–239.
36. Smadja DM, Bieche I, Uzan G, et al. PAR-1 activation on human late endothelial 18. Smadja DM, Bieche I, Silvestre JS, et al. Bone morphogenetic proteins 2 and 4 progenitor cel s enhances angiogenesis in vitro with upregulation of the SDF- are selectively expressed by late outgrowth endothelial progenitor cel s and pro- 1/CXCR4 system. Arterioscler Thromb Vasc Biol 2005; 25: 2321–2327.
mote neoangiogenesis. Arterioscler Thromb Vasc Biol 2008; 28: 2137–2143.
37. Gehling UM, Ergun S, Schumacher U, et al. In vitro differentiation of endothe- 19. Wojakowski W, Tendera M, Kucia M, et al. Mobilisation of bone marrow-de- lial cel s from AC133-positive progenitor cel s. Blood 2000; 95: 3106–3112.
rived Oct-4+ SSEA-4+ very small embryonic-like stem cel s in patients with 38. Kawamoto A, Iwasaki H, Kusano K, et al. CD34-positive cel s exhibit increased acute myocardial infarction. J Am Coll Cardiol 2009; 53: 1–9.
potency and safety for therapeutic neovascularisation after myocardial infarc- 20. Pittenger MF, Mackay AM, Beck SC, et al. Multilineage potential of adult human tion compared with total mononuclear cel s. Circulation 2006; 114: 2163–2169.
mesenchymal stem cel s. Science 1999; 284: 143–147.
39. Kocher AA, Schuster MD, Szabolcs MJ, et al. Neovascularisation of ischaemic 21. Smadja DM, Guerin CL, Boscolo E, et al. alpha6-Integrin is required for the ad- myocardium by human bone-marrow-derived angioblasts prevents cardiomyo- hesion and vasculogenic potential of hemangioma stem cel s. Stem Cel s 2014; cyte apoptosis, reduces remodeling and improves cardiac function. Nat Med 32: 684–693.
2001; 7: 430–436.
22. Melero-Martin JM, De Obaldia ME, Kang SY, et al. Engineering robust and 40. Barcelos LS, Duplaa C, Krankel N, et al. Human CD133+ progenitor cel s pro- functional vascular networks in vivo with human adult and cord blood-derived mote the healing of diabetic ischaemic ulcers by paracrine stimulation of angio- progenitor cel s. Circ Res 2008; 103: 194–202.
genesis and activation of Wnt signaling. Circ Res 2009; 104: 1095–1102.
23. Tateishi-Yuyama E, Matsubara H, Murohara T, et al. Therapeutic angiogenesis 41. Shin DM, Zuba-Surma EK, Wu W, et al. Novel epigenetic mechanisms that con- for patients with limb ischaemia by autologous transplantation of bone-marrow trol pluripotency and quiescence of adult bone marrow-derived Oct4 (+) very cel s: a pilot study and a randomised controlled trial. Lancet 2002; 360: 427–435.
small embryonic-like stem cel s. Leukemia 2009; 23: 2042–2051.
24. Guven H, Shepherd RM, Bach RG, et al. The number of endothelial progenitor 42. Havens AM, Sun H, Shiozawa Y, et al. Human and murine very small em- cell colonies in the blood is increased in patients with angiographically signifi- bryonic-like cel s represent multipotent tissue progenitors, in vitro and in vivo. cant coronary artery disease. J Am Coll Cardiol 2006; 48: 1579–1587.
Stem Cel s Dev 2014; 23: 689–701.
25. Massa M, Campanelli R, Bonetti E, et al. Rapid and large increase of the fre- 43. Zuba-Surma EK, Guo Y, Taher H, et al. Transplantation of expanded bone mar- quency of circulating endothelial colony-forming cel s (ECFCs) generating late row-derived very small embryonic-like stem cel s (VSEL-SCs) improves left outgrowth endothelial cel s in patients with acute myocardial infarction. Exp ventricular function and remodelling after myocardial infarction. J Cell Mol Hematol 2009; 37: 8–9.
Med 2011; 15: 1319–1328.
26. Toshner M, Voswinckel R, Southwood M, et al. Evidence of dysfunction of en- 44. Al-Khaldi A, Al-Sabti H, Galipeau J, et al. Therapeutic angiogenesis using auto- dothelial progenitors in pulmonary arterial hypertension. Am J Respir Crit Care logous bone marrow stromal cel s: improved blood flow in a chronic limb is- Med 2009; 180: 780–787.
chaemia model. Ann Thorac Surg 2003; 75: 204–209.
27. Smadja DM, Mauge L, Gaussem P, et al. Treprostinil increases the number and 45. Al-Khaldi A, Eliopoulos N, Martineau D, et al. Postnatal bone marrow stromal angiogenic potential of endothelial progenitor cel s in children with pulmonary cel s elicit a potent VEGF-dependent neoangiogenic response in vivo. Gene hypertension. Angiogenesis 2011; 14: 17–27.
Ther 2003; 10: 621–629.
28. Smadja DM, Mauge L, Nunes H, et al. Imbalance of circulating endothelial cel s 46. Kucia M, Reca R, Campbell FR, et al. A population of very small embryonic-like and progenitors in idiopathic pulmonary fibrosis. Angiogenesis 2013; 16: (VSEL) CXCR4 (+)SSEA-1 (+)Oct-4+ stem cel s identified in adult bone mar- row. Leukemia 2006; 20: 857–869.
29. Fadini GP, Sartore S, Albiero M, et al. Number and function of endothelial pro- 47. Kucia M, Zuba-Surma E, Wysoczynski M, et al. Physiological and pathological genitor cel s as a marker of severity for diabetic vasculopathy. Arterioscler consequences of identification of very small embryonic like (VSEL) stem cel s in Thromb Vasc Biol 2006; 26: 2140–2146.
adult bone marrow. J Physiol Pharmacol 2006; 57 (Suppl 5): 5–18.
30. Ingram DA, Mead LE, Tanaka H, et al. Identification of a novel hierarchy of en- 48. Paczkowska E, Kucia M, Koziarska D, et al. Clinical evidence that very small dothelial progenitor cel s using human peripheral and umbilical cord blood. embryonic-like stem cel s are mobilized into peripheral blood in patients after Blood 2004; 104: 2752–2760.
stroke. Stroke 2009; 40: 1237–1244.
31. Smadja DM, Bieche I, Emmerich J, et al. PAR-1 activation has different effects on the angiogenic activity of endothelial progenitor cel s derived from human adult and cord blood. J Thromb Haemost 2006; 4: 2729–2731.
Schattauer 2015 Thrombosis and Haemostasis 113.5/2015 Downloaded from www.thrombosis-online.com on 2015-01-26 ID: 1000471449 IP: 136.165.97.221 Note: Uncorrected proof, epub ahead of print online For personal or educational use only. No other uses without permission. All rights reserved.

Source: http://stemcells-project.eu/public/www/files/user/Bone-marrow-derived%20very%20small%20embryonic-like%20stem%20cells%20in%20patients%20evidence%20of%20vasculogenic%20potential.pdf