Human CD34 /KDR Cells Are Generated From Circulating CD34

Human CD34�/KDR� Cells Are Generated FromCirculating CD34� Cells After Immobilization on

Activated PlateletsH.C. de Boer, M.M. Hovens, A.M. van Oeveren-Rietdijk, J.D. Snoep, E.J.P. de Koning, J.T. Tamsma,

M.V. Huisman, A.J. Rabelink, A.J. van Zonneveld

Objective—The presence of kinase-insert domain-containing receptor (KDR) on circulating CD34� cells is assumed to beindicative for the potential of these cells to support vascular maintenance and repair. However, in bone marrow and ingranulocyte colony-stimulating factor (G-CSF)-mobilized peripheral blood, less than 0.5% of CD34� cells co-expressKDR. Therefore, we studied whether CD34�/KDR� cells are generated in the peripheral circulation.

Methods and Results—Using an ex vivo flow model, we show that activated platelets enable CD34� cells to home to sitesof vascular injury and that upon immobilization, KDR is translocated from an endosomal compartment to thecell-surface within 15 minutes. In patients with diabetes mellitus type 2, the percentage of circulating CD34�

co-expressing KDR was significantly elevated compared to age-matched controls. When treated with aspirin, thepatients showed a 49% reduction in the generation of CD34�/KDR� cells, indicating that the level of circulatingCD34�/KDR� cells also relates to in vivo platelet activation.

Conclusion—Circulating CD34�/KDR� are not mobilized from bone marrow as a predestined endothelial progenitor cellpopulation but are mostly generated from circulating multipotent CD34� cells at sites of vascular injury. Therefore, thenumber of circulating CD34�/KDR� cells may serve as a marker for vascular injury. (Arterioscler Thromb Vasc Biol.2011;31:408-415.)

Key Words: aspirin � blood flow � diabetes mellitus � platelets � vascular biology

Numerous reports have shown an inverse correlationbetween the number of circulating CD34�/kinase-insert

domain-containing receptor (KDR)� cells and hemodynamicand metabolic cardiovascular risk factors, such as hyperten-sion, hyperglycemia, hypercholesterolemia, and end-stagerenal disease.1 These circulating progenitor cells are consid-ered to be a bone marrow (BM)–derived subpopulation of theCD34-positive hematopoietic stem cell fraction that coex-press vascular endothelial growth factor (VEGF) receptortype-2/KDR.2 Because KDR expression is crucial for vascu-lar development3 and isolated CD34�/KDR� cells may con-tribute to reendothelialization,4,5 these cells are considered toplay a role as endothelial progenitor cells in vascular main-tenance and repair.

See accompanying article on page 243Consequently, a reduced number of circulating CD34�/

KDR� cells was proposed to be a causal factor in thedevelopment of endothelial dysfunction and atherosclerosis.6

Recently, we showed that hemostatic factors deposited at the

site of an injury, such as activated platelets, platelet products,and fibrin, support the homing of human CD34� cells underflow.7 The hemostatic clot provides local signals, such asP-selectin, stromal cell-derived factor-1/CXC-chemokineligand-12 (SDF-1/CXCL12), and VEGF, which facilitatedthe arrest and survival of CD34� cells and induced theircommitment toward an endothelial cell (EC)–like pheno-type.7 Given the importance of VEGF in endothelial differ-entiation, we investigated the dynamics of KDR expressionby CD34� cells that have adhered to platelet-rich sites of ECactivation under conditions of flow. Surprisingly, we ob-served that most, if not all, CD34� cells exhibit KDRexpression within 15 minutes of shear exposure. Therefore,CD34�/KDR� cells may not be mobilized from BM as aseparate progenitor lineage but are predominantly generatedfrom CD34� cells at sites of vascular injury. To address thishypothesis, we conducted a detailed ex vivo study of the earlyphenotypic changes that CD34� cells undergo after homing.

We demonstrate that CD34� cells can rapidly translocateKDR from intracellular storage organelles to the cell surface

Received on: May 20, 2010; final version accepted on: October 16, 2010.From the Departments of Nephrology (H.C.dB., A.M.vO-R., E.J.P.dK., A.J.R., A.J.vZ), General Internal Medicine and Endocrinology (Section of

Vascular Medicine) (M.M.H., J.D.S., J.T.T., M.V.H.), and Clinical Epidemiology (J.D.S.), and the Einthoven Laboratory for Experimental VascularMedicine (H.C.dB., A.M.vO-R., A.J.R., A.J.vZ.), all at the Leiden University Medical Center, Leiden, The Netherlands; the Department of InternalMedicine (M.M.H.), Rijnstate Hospital, Arnhem, the Netherlands.

Correspondence to A.J. van Zonneveld, Department of Nephrology, Leiden University Medical Center, Albinusdreef 2, 2333 ZA Leiden, theNetherlands. E-mail [email protected]

© 2011 American Heart Association, Inc.

Arterioscler Thromb Vasc Biol is available at http://atvb.ahajournals.org DOI: 10.1161/ATVBAHA.110.216879

408

by guest on March 17, 2018

http://atvb.ahajournals.org/D

ownloaded from

by guest on M

arch 17, 2018http://atvb.ahajournals.org/

Dow

nloaded from



ownloaded from

by guest on M


Dow

nloaded from



ownloaded from

by guest on M


Dow

nloaded from



ownloaded from

by guest on M


Dow

nloaded from



ownloaded from

by guest on M


Dow

nloaded from

http://atvb.ahajournals.org/










in a regulated phosphatidylinositol-3-kinase (PI3K)–depen-dent pathway. Then, extensive analysis of the percentage ofCD34� cells coexpressing KDR in BM, umbilical cord blood(UCB), and peripheral blood (PB) after G-CSF mobilization,in PB of control subjects and in vascular injury–pronepatients with diabetese mellitus type 2 (DM2), also stronglysupports the in vivo generation of CD34�/KDR� cells in theperiphery. Finally, to assess the role of platelets in theperipheral generation of CD34�/KDR� cells in patients withDM2, we investigated the effect of aspirin (300 mg/d) in aplacebo-controlled crossover study.

MethodsThe supplemental material (available online at http://atvb.ahajournals.org) provides a full description of all methods.

PerfusionsPerfusions of CD34� cells were performed as previously described.7

Subjects and Study DesignCirculating cells were studied in different samples: human BM,UCB, PB of G-CSF–stimulated healthy donors, PB of patients withDM2, and PB of age-matched control subjects. Subjects with DMwere recruited from general practitioners affiliated with LeidenUniversity Medical Center, Leiden, the Netherlands. Details aboutthe randomized placebo-controlled study on the effect of aspirin insubjects with DM2 have been previously published.8

Results

CD34� Cells Homed to an EC Monolayer UnderShear Express KDR on Their SurfaceBecause platelet-derived VEGF signaling through KDR isone of the most likely drivers of EC differentiation3 invasculogenesis, we studied the dynamics of KDR expressionby CD34� cells. On freshly isolated nonsheared CD34�

(CD45dim) cells, KDR expression was not detectable aboveisotype (IT) levels (Figure 1A). When these cells wereperfused in the ex vivo injury model, nonattaching CD34�

cells (non) still lacked KDR expression, whereas recoveredCD34� cells, which had been subjected to shear for 15minutes (att, Figure 1A) showed a significant surface expres-sion of KDR (P�0.042). As a control, both CD34 andP-selectin glycoprotein ligand-1 (PSGL-1) expression onnonsheared cells was confirmed (Figure 1B [P�0.001] andFigure 1C [P�0.01], respectively) but was not altered oncells subjected to shear. KDR upregulation was observed onalmost all recovered/sheared cells, as evident in the histogramin Figure 1F, which shows that the entire peak has shifted tothe right compared with the level of IT-incubated CD34�

cells (dashed line as a reference). For comparison, histogramsare shown for KDR expression of nonperfused CD34� cells(Figure 1D) and nonattached cells after perfusion (Figure 1E).Immunohistochemical staining of the EC-adherent shear-subjected CD34� cells (Figure 1H) confirmed homogeneousexpression of KDR on the cell surface (Figure 1G).

CD34� Cells Harbor KDR in Intracellular PoolsAfter firm adhesion of CD34� to activated ECs, KDR wasdetectable after only 15 minutes of shear, suggesting therelease from an intracellular storage pool rather than de novosynthesis. To demonstrate this, freshly isolated CD34� cellswere fixed, permeabilized or not, and stained for the extra-cellular domain of KDR and CD34. IT-matched antibodiesdetermined background levels. Indeed, KDR expression onnaı̈ve cells was only detectable after permeabilization (Figure1I, perm, P�0.01). In contrast, CD34 was not differentiallyexpressed on permeabilized or nonpermeabilized cells (datanot shown).

KDR Resides With CXC-Chemokine Receptor-4(CXCR4 or SDF-receptor) in an EarlyEndosomal CompartmentHuman CD34� cells harbor intracellular CXCR4 in an earlyendosomal compartment, which can be functionally ex-pressed to mediate SDF-1–dependent homing.9 To investi-gate whether KDR resides in similar structures, we studiedcolocalization of both KDR and CXCR4 with the earlyendosome marker-1 (EEA-1) in CD34� cells. Confocal laserscanning microscopy of permeabilized cells showed colocal-ization of KDR and EEA-1 (supplemental Figure, A).CXCR4 showed similar colocalization with EEA-1 in perme-abilized CD34� cells (supplemental Figure, B). Costainingfor KDR and CXCR4 confirmed colocalization in the endo-somal compartment (supplemental Figure, C). IT controlswere negative on both permeabilized (supplemental Figure,F) and nonpermeabilized (supplemental Figure, D, E, and G)cells.

Homed CD34� Cells Mobilize KDR in aPI3K-Dependent FashionOn homing, CD34 cells are subjected to the injury-associatedmicroenvironment, containing activated platelets and fibrin(Figure 2A and 2B). Platelet-derived factors, in particularSDF-1, can signal through CXCR4 to initiate signal trans-duction pathways that prepare the cell for migration anddifferentiation. A common denominator of these pathways isthe downstream activation of PI3K.10 To investigate whetherthe translocation of KDR was the result of a PI3K-specificevent, we visualized PI3K-dependent activation in these cells.First, adherent CD34� cells (Figure 2D) exhibited denseareas of polymerized filamentous actin (F-actin) in thecortical cytosol, demonstrating the activated status of thecells, which is secondary to PI3K-dependent activation in-volving Rac1, protein kinase C-�, extracellular signal–regu-lated kinase 1/2, and WAVE2.11 In contrast, nonshearedCD34� cells show diffuse low-level staining of F-actin(Figure 2C).

A second PI3K-dependent process that follows homingis the phosphorylation of Akt/PKB (p-Akt, protein kinaseB).12 Adherent cells showed a clear signal for p-Akt(Figure 2E-2G) as opposed to the nonperfused control cells(Figure 2C). The presence of membrane-associated p-Aktat the contact sites of CD34� cells and ECs (Figure 2F and

de Boer et al Circulating CD34�/KDR� Cells Reflect Vascular Injury 409



ownloaded from

http://atvb.ahajournals.org

http://atvb.ahajournals.org


2G, different angles) again confirms that the CD34� cellsare rapidly activated on firm adhesion. Next, we assessedwhether the KDR translocation from the intracellularcompartment to the cell surface is also dependent on PI3Kactivation. To that end, CD34� cells were preincubatedwith the PI3K inhibitor wortmannin (WM, 100 nmol/L)before and during perfusion, adherent cells were retrieved,and KDR expression was assessed by fluorescence-acti-vated cell sorter (FACS). Inhibition of PI3K by WMshowed a significantly reduced KDR signal (Figure 1J;�WM versus �WM; P�0.03), whereas the signal ob-tained with the IT was not affected. The specific KDRexpression, calculated by subtracting IT signals from theKDR signals (KDR-IT), revealed a decrease of 75%(P�0.01).

BM, UCB, or G-CSF–Mobilized PB Contains LowLevels of CD34�/KDR�

Thus far, we demonstrated that few freshly isolated CD34�

cells express KDR. We reasoned that CD34�/KDR� cellsmight be generated in vivo at platelet-rich sites of vascularinjury. This hypothesis predicts that the CD34� cell popula-tion, freshly recruited to the peripheral circulation of healthysubjects by mobilizing agents such as G-CSF, would containlow numbers of cells coexpressing KDR. In contrast, inpatients with endothelial or ischemic injury, the percentage ofCD34� cells coexpressing KDR would increase because of invivo activation of the cells on the injured platelet-rich ECsurface. To validate this hypothesis, we used a detailedquantitative FACS analysis of CD34� and CD34�/KDR�

cells. The FACS analysis was performed using a multipara-

Figure 1. A through C, Shear-subjected CD34� cells exhibit rapid surface expression of KDR originating from an intracellular pool. Iso-lated CD34� cells were perfused over procoagulant ECs, covered with fibrin-rich thrombi. Then, KDR (A), CD34 (B), and PSGL-1 (C)surface expression levels were determined by FACS (expressed as mean fluorescence intensity) on nonsheared or sheared CD34�

cells, either nonattached (non) or attached and recovered (att). Background levels were determined using IT controls. D through F, His-tograms of gated cells (CD34pos and CD45dim) expressing KDR are shown for nonperfused cells (D), perfused/nonattached cells (E),and perfused/attached cells (F). IT binding is shown as background level (vertical dashed line). G and H, KDR expression on perfusedand attached CD34� cells was analyzed using confocal laser scanning microscopy and showed homogeneous KDR expression on thecell surface (G), colocalizing with CD34 staining (H). Staining is combined with Nomarski projection (scale bar, 5 �m). I, To show thatKDR is localized in an intracellular pool, CD34� cells were fixed, either not permeabilized (nonperm) or permeabilized (perm), andstained for expression of KDR. IT controls were used to determine background staining. J, KDR expression on shear-subjected CD34�

cells was reduced (P�0.03) in the presence of the PI3K inhibitor WM (�WM vs �WM); this did not affect the MFI of IT controls. Sub-traction of the IT signal (KDR-IT) from the KDR signal showed a reduction of KDR upregulation of 75% (P�0.01). In I, the overall MFIlevels of KDR-staining and IT controls were higher vs those in A and J, probably because of fixation of the cells.

410 Arterioscler Thromb Vasc Biol February 2011



ownloaded from


metric gating strategy, including fluorescent microbeads, thatpermits the acquisition of absolute numbers of cells pervolume of blood. However, because UCB is anticoagulatedwith a fixed amount of 8 mL of sodium citrate, the dilution ofUCB samples may differ depending on the volume collected.Therefore, in this case, we expressed the number of CD34�

cells per 103 CD45� leukocytes. BM samples contained5.14�1.06 CD34� cells/103 CD45� leukocytes (Figure 3,left axis), of which 0.34%�0.20% were CD34�/KDR� (rightaxis). UCB samples contained 3.42�1.35 CD34� cells/103

CD45� leukocytes, of which only 0.23%�0.09% coex-pressed KDR. Next, we assessed that PB of control subjects(n�15) contained 0.30�0.03 CD34� cells/103 leukocytes, ofwhich 0.72%�0.19% were CD34�/KDR�. As expected, thenumber of CD34� cells in G-CSF–mobilized PB of healthysubjects (n�4) was largely increased, amounting to2.27�0.52 CD34� cells/103 CD45� leukocytes, which ac-counts for a 7.5-fold increase of CD34� cells. Of these cells,only 0.19%�0.18% coexpressed KDR. Data provided asmean�SEM.

Percentage of Circulating CD34� CellsCoexpressing KDR Is Elevated in PatientsWith DM2Patients with DM2 may well display a vascular phenotypethat is uniquely suited to study the correlation betweenvascular injury and circulating CD34�/KDR� cells. Thepathophysiological mechanisms involved in this chronic vas-cular disease include EC dysfunction,13 ongoing vascularinjury,14 and increased platelet aggregability.15 To investigatewhether the number of circulating CD34�/KDR� cells inpatients at risk for chronic vascular disease relates to endo-thelial injury and platelet activation, we performed a random-ized, double-blind, placebo-controlled crossover trial in pa-tients with DM2. All subjects received placebo and aspirin(300 mg/d; n�20) for 6 weeks. The control subjects wereoriginally included for a study on metabolic syndrome, forwhich hypertension, waist circumference, fasting blood glu-cose, triglyceride, and high-density lipoprotein cholesterollevels were defining variables. However, the control subjectsonly had 1 or 2 of these traits, whereas 3 or more traits areneeded to define them as positive for metabolic syndrome.16

Therefore, these subjects are considered excellent controls forthe patients with diabetes, who had comparable ages andanthropometric indexes.

Subject characteristics are summarized in the Table, andstatistical differences between the groups are shown with anasterisk. There were no differences in age, sex, and bloodpressure (either systolic or diastolic). Patients with DM2 hada higher body mass index and waist circumference andelevated levels of fasting blood glucose, homeostasis modelassessment of insulin resistance, glycohemoglobin, andC-reactive protein, indicating a low-grade inflammatory state.None of the statistical differences between groups disap-peared when we restricted the analysis to male subjects withDM2 only (data not shown), indicating that these differencesare consistent with the diabetic profile. Circulating cells were

Figure 2. Platelet thrombi support rolling and firm adhesion ofCD34� to an EC monolayer. A and B, Scanning electron micro-graph (bar, 5 �m) (A) and light microscopic image (bar, 50 �m)(B) of adherent CD34� cells downstream of platelet aggregates(plts), enforced with fibrin (plts�fibrin), deposited on ECs afterflow. C through G, Confocal laser scanning microscopy of non-sheared (C) and sheared (D-G) CD34� cells stained for F-actin(C-G; red fluorescence) and p-Akt (C and E-G, blue fluores-cence), demonstrating shear-induced polymerization of F-actinand flattening of the cells (D, dashed outline), leading to anincrease in diameter of the adhered cells. In E and F, shearedCD34� cells and nonsheared cells (C) were stained forF-actin (red) and p-Akt (blue), demonstrating the presence ofmembrane-associated p-Akt at the contact sites of CD34� cellsand ECs in an X-Y projection (E) and 2 z planes (F and G) visu-alized from different angles. The asterisk represents z planes.

Figure 3. The percentage of circulating CD34�/KDR� cells perCD34� cells progresses with diabetic conditions associated withvascular injury and platelet activation. The number of CD34�

cells (left axis, white bars) was measured (expressed as numberper 103 CD45� leukocytes) by FACS in BM, UCB, and PB afterG-CSF treatment (cntrls�G-CSF), in PB from patients with DM2,and in age-matched controls (cntrls). The percentage of CD34�/KDR� cells per CD34� cell fraction was calculated (right axis,gray bars).




ownloaded from


assessed using the FACS assay with microbeads, permittingthe acquisition of absolute counts of cells, expressed asnumber per milliliter of whole blood. No significant differ-ences were observed in the number of CD34� cells at the endof the aspirin treatment period compared with placebo (Fig-ure 4A: 1199�267 versus 1015�230 cells/mL; P�0.359). Incontrast, the number of circulating CD34�/KDR� cells wassignificantly reduced after aspirin treatment, lowering the cellcounts by 47%, from 32.5�5.7/mL to 17.1�5.3/mL(P�0.046, Figure 4B). In agreement with our hypothesis, thepercentage of circulating CD34� cells coexpressing KDR� inthe patients with DM2 at baseline was markedly elevatedcompared with the control group: in patients with DM2,3.5%�0.8% of the circulating CD34� cells coexpressedKDR, which is 5-fold higher than the 0.7%�0.2% in thecontrol group (P�0.007, Figure 3).

CD34�/KDR� Cell Numbers Are Correlated WithPlasma Levels of Soluble P-SelectinTo further support the role of platelets in the generation ofCD34�/KDR� in vivo, we compared the levels of solubleP-selectin (sP-sel) as a marker for in vivo platelet activation17

in the placebo and aspirin-treated patients. Aspirin treatmentdid not affect the average values of sP-sel between the placeboand aspirin-treated groups (51.1�4.9 versus 54.3�6.0 ng/mL,respectively; P�0.437; Figure 4C). However, when lookingat the subjects individually, 50% of the patients showed a

reduction in sP-sel, whereas the other half showed increasedsP-sel levels, confirming previous observations that the sen-sitivity for aspirin and, thus, the inhibition of in vivo basalplatelet reactivity, in patients with DM2 is highly subjectspecific.18 To investigate the relation between effective plate-let aggregation inhibition and circulating cells, we calculatedthe increase or decrease (� values) of sP-sel, CD34�, andCD34�/KDR� cells for the individual patients and performeda linear regression analysis. As expected, changes in thenumber of CD34� cells did not correlate with changes in thelevel of sP-sel (P�0.206, r2�0.105, data not shown), indi-cating that CD34� cell numbers are not related to the degreeof in vivo platelet activation. In contrast, � levels of CD34�/KDR� cells were strongly correlated with � levels of sP-selafter aspirin treatment (Figure 4D, P�0.001, r2�0.589).Taken together, elevated platelet activation in patients withDM2 is strongly correlated with an increased number ofcirculating CD34�/KDR� cells. Conversely, when plateletactivation is inhibited, as in the “aspirin-responsive” patientswith DM2, the number of circulating CD34�/KDR� cells isreduced accordingly.

DiscussionOur study supports a novel concept for the origin of circu-lating CD34�/KDR� cells (Figure 5). In contrast to conceptsthat CD34�/KDR� cells are mobilized from the BM as aunique predestined lineage with an endothelial fate, our datasuggest that these cells more likely reflect the plasticity ofCD34� cells and their capacity to respond to environmentalcues. First, we determined that after G-CSF treatment, only

Table. Subject Characteristics at Baseline*

CharacteristicsControls(n�15)

Patients With DM2(n�20)

Age, y 59.3�1.6 54.5�2.1

Female sex† 0 (0) 5 (25)

BMI, kg/m2 26.9�0.4 31.5�1.2‡

Waist circumference, cm 101�1 106�2§

Tension, mm Hg

Systolic 146�4 150�4

Diastolic 85�2 90�2

Smoking† 0 (0) 2 (10)

MetS† 0 (0) 18 (90)¶

Fasting blood glucose, mmol/L 5.0�0.1 7.7�0.2¶

HbA1c, % 4.9�0.1 5.9�0.2¶

HOMA-IR 1.1�0.1 2.3�0.4‡

Total cholesterol, mmol/L 5.7�0.2 5.5�0.2

HDL cholesterol, mmol/L 1.2�0.1 1.4�0.1

LDL cholesterol, mmol/L 3.3�0.1 3.8�0.2

Triglycerides, mmol/L 2.5�0.4 2.0�0.3

Creatinine, �mol/L 94.3�3.3 85.1�3.2

CRP, mg/L 1.1�0.2 8.9�2.1¶

BMI indicates body mass index; CRP, C-reactive protein; HbA1c, glycatedhemoglobin; HDL, high-density lipoprotein; HOMA-IR, homeostasis modelassessment of insulin resistance; LDL, low-density lipoprotein; MetS, metabolicsyndrome.

*Data are given as mean�SEM unless otherwise indicated.†Data are given as number (percentage) of each group.‡P�0.01. §P�0.05. ¶P�0.001.

Figure 4. Assessment of absolute and � values of circulatingcells and sP-sel. A through C, Absolute numbers are shown ofCD34� cells (number/mL) (A), CD34�/KDR� cells (number/mL)(B), or sP-sel (ng/mL) (C) in the placebo or aspirin-treatedperiod. Mean�SEM levels are presented with bars, and resultsper subject are shown, connecting placebo- with aspirin-obtained values. The probability values are calculated compar-ing placebo with aspirin. D, Linear regression is shown between� values of sP-sel and CD34�/KDR� cells.




ownloaded from


0.19%�0.18% of the mobilized peripheral CD34� cellscoexpress KDR. This level of expression is similar to that ofthe CD34� population in BM, which amounts to0.34%�0.19% and confirms that during mobilization, little“activation” of these cells occurs. Second, we demonstratedthat most, if not all, CD34� cells can rapidly upregulate KDRexpression once recruited to platelet- and fibrin-rich ECsurfaces. Third, in patients with DM2 who are known todisplay ongoing endothelial injury with associated plateletaggregation, we observed a 5-fold increase in the percentageof circulating CD34� cells coexpressing KDR compared withcontrol subjects. Moreover, treatment with aspirin (300 mg/d)reduced the number of circulating CD34�/KDR� cells by47% in these patients, concomitant with markers of systemicplatelet activation. Together, these data support our hypoth-esis that, also in vivo, CD34�/KDR� cells are generated fromCD34� cells at sites of vascular injury. We propose that mostof the CD34�/KDR� cells will remain associated with thevascular wall to participate in vascular repair (Figure 5; arrowa), whereas a fraction of these cells may reenter the circula-tion from the injured vascular wall secondary to shear ormechanical stimuli (Figure 5, arrow b), thereby reflecting therate of conversion of CD34� cells at platelet-rich sites ofvascular injury. Thus, the number of CD34�/KDR� cellsmay provide an index of systemic vascular injury.

Interestingly, in conditions of acute vascular ischemia andEC injury, an increase in the number of circulating CD34�/KDR� cells has been observed (eg, thermal injury,19

radiation-induced endothelial injury,20 sepsis,21 acute ische-mic stroke,22 acute myocardial infarction,23 and even briefexposure to second-hand smoke24). Some of these studiesreported that the acute elevation of circulating CD34�/KDR�

cells was associated with a concomitant increase in levels ofmobilizing factors, such as VEGF and SDF-1, suggesting acausal relationship.21,25 Furthermore, markers of endothelialdamage/dysfunction (ie, von Willebrand factor and solubleE-selectin) and the platelet marker sP-sel26 were reported tobe elevated, suggesting that these acute conditions are asso-ciated with in vivo vascular injury and platelet activation.These data are in agreement with our concept that the

CD34�/KDR� cell fraction does not increase as a result ofmobilization from BM but may rather be generated in theperiphery from CD34� cells immobilized at platelet-rich sitesof vascular injury.

We demonstrated that the absolute number of CD34�

cells/per 103 CD45� leukocytes was significantly reduced inpatients with DM2 compared with controls (Figure 3). This isin line with many studies that have consistently demonstrateda reduction of these cells in patients with DM2 and in otherpatient groups with an increased risk for chronic vasculardisease.27

In contrast, we show that the percentage of CD34� cellscoexpressing KDR is increased in patients with DM2; andthis paralleled absolute numbers of these cells (data notshown). This seems contradictive to published results; how-ever, in many reports, patients received aspirin27 or plateletaggregation inhibitors,28 and this may explain why thesearticles document lower amounts of CD34�/KDR� cells intheir patient cohorts.

The link between CD34�/KDR� cell numbers and plateletactivation in the patients with DM2 was only revealed whenwe distinguished patients who responded to aspirin fromnonresponders. This may be explained by the fact that, inpatients with coronary artery disease, diabetic patients have ahigh prevalence of laboratory-defined aspirin resistance.18

Our ex vivo studies revealed that surface expression ofKDR is already detectable within 15 minutes after adhesionto platelet-rich EC surfaces. This short time span indicatesthat KDR mobilization is not mediated by increased tran-scription, as was proposed in another model in which CD34�

cells were subjected to shear for 24 hours.29 Indeed, wedetected an intracellular pool of KDR, associated with anEEA compartment, in which CXCR4 has also been local-ized.9 Like other growth factor receptors, KDR30 andCXCR49 are subject to a low constitutive rate of endocytosis.This internalization involves the formation of endocyticvesicles from clathrin-coated pits that subsequently fuse withearly (EEA-1–positive) endosomes, from which the receptorscan either traffic to late endosomes and the lysosomalcompartment for degradation or recycle to the cell mem-brane.31 This constitutive downregulation of growth factorreceptors may serve 2 important purposes. First, it desensi-tizes the cell to avoid premature activation by growth factors.In particular, for CD34� stem cells, this may be essential tomaintain plasticity during their trafficking from and towardBM niches. Second, when cells are recruited to sites ofinflammation or tissue repair, targeted recycling is essentialto generate polarity of the cells. Active recycling of KDR andCXCR4 to the leading edge may sensitize the cell to VEGF-and SDF-1–induced chemotactic signals.32 VEGF and SDF-1play important roles as chemoattractants in progenitor cellmobilization and homing, and both factors may also serve asEC differentiation factors.33,34 Because VEGF and SDF-1 areactively released from activated platelets, the rapid upregu-lation of KDR and CXCR4 by CD34� cells that have adheredto platelet-rich sites of EC injury may well further stimulatethese cells to enter an EC fate.7 The fact that WM inhibitedKDR expression indicates that recycling of this receptor is theresult of PI3K-dependent signaling pathway(s) that follow on

Figure 5. Schematic representation of platelet-dependent gen-eration of CD34�/KDR� cells from CD34� cells. Freshly mobi-lized CD34� cells hardly express KDR. On rolling/arrest to plate-let-rich sites of vascular injury and exposure to shear, KDR israpidly translocated from intracellular storage organelles to thecell surface. Activated CD34� cells can either progress to vas-cular progenitor cells (arrow a) or dissociate (arrow b) and con-stitute the circulating pool of CD34�/KDR� cells.




ownloaded from


homing-mediated activation. It was previously demonstratedthat ligation of PSGL-1 with a soluble form of P-selectinactivated expression of the integrins MAC-1 (CD11b/CD18)and �v�5.7 Therefore, integrin-mediated “outside-in-signaling” could be one mechanism responsible for theelevation of PI3K.35 In addition, PSGL-1-ligation itself canlead to PI3K activation in an Src kinase–dependent path-way.10 A third pathway that may lead to PI3K activation isexposure of the adherent CD34� cells to SDF-1 expressed byactivated platelets.36 By using FACS analysis, we determinedthat low levels of CXCR4 are present on freshly isolatedCD34� cells (data not shown); on binding of SDF-1, the cellscan be activated to a promigratory state in a PI3K-dependentfashion.37

Finally, CD34� cell fractions have been administered inclinical trials aimed to develop cell-based strategies to treatpatients with cardiac and peripheral ischemic disease.38 If,indeed, CD34�/KDR� cells represent the provasculogeniccomponent in these studies, the therapeutic potential of theCD34� cells may have been underestimated because theCD34�/KDR� cell fraction and the CD34�/KDR� cells,which compose �99% of the circulating stem cells in healthyindividuals, may participate in a provasculogenic response.

Sources of FundingThis work was supported in part by grant NHS2005B106 from theNetherlands Heart Foundation (Dr H.C. de Boer); and the TeRMSmart Mix Program of the Netherlands Ministry of Economic Affairsand the Netherlands Ministry of Education, Culture and Science (DrA.J. van Zonneveld).

DisclosuresNone.

References1. Werner N, Nickenig G. Influence of cardiovascular risk factors on endo-

thelial progenitor cells: limitations for therapy? Arterioscler Thromb VascBiol. 2006;26:257–266.

2. Asahara T, Takahashi T, Masuda H, Kalka C, Chen D, Iwaguro H, InaiY, Silver M, Isner JM. VEGF contributes to postnatal neovascularizationby mobilizing bone marrow-derived endothelial progenitor cells. EMBOJ. 1999;18:3964–3972.

3. Sakurai Y, Ohgimoto K, Kataoka Y, Yoshida N, Shibuya M. Essentialrole of Flk-1 (VEGF receptor 2) tyrosine residue 1173 in vasculogenesisin mice. Proc Natl Acad Sci U S A. 2005;102:1076–1081.

4. Peichev M, Naiyer AJ, Pereira D, Zhu Z, Lane WJ, Williams M, Oz MC,Hicklin DJ, Witte L, Moore MA, Rafii S. Expression of VEGFR-2 andAC133 by circulating human CD34(�) cells identifies a population offunctional endothelial precursors. Blood. 2000;95:952–958.

5. Pelosi E, Valtieri M, Coppola S, Botta R, Gabbianelli M, Lulli V,Marziali G, Masella B, Muller R, Sgadari C, Testa U, Bonanno G,Peschle C. Identification of the hemangioblast in postnatal life. Blood.2002;100:3203–3208.

6. Dimmeler S, Zeiher AM. Vascular repair by circulating endothelial pro-genitor cells: the missing link in atherosclerosis? J Mol Med. 2004;82:671–677.

7. de Boer HC, Verseyden C, Ulfman LH, Zwaginga JJ, Bot I, Biessen EA,Rabelink TJ, van Zonneveld AJ. Fibrin and activated platelets coopera-tively guide stem cells to a vascular injury and promote differentiationtowards an endothelial cell phenotype. Arterioscler Thromb Vasc Biol.2006;26:1653–1659.

8. Hovens MM, Snoep JD, Groeneveld Y, Frolich M, Tamsma JT, HuismanMV. Effects of aspirin on serum C-reactive protein and interleukin-6

levels in patients with type 2 diabetes without cardiovascular disease: arandomized placebo-controlled crossover trial. Diabetes Obes Metab.2008;10:668–674.

9. Zhang Y, Foudi A, Geay JF, Berthebaud M, Buet D, Jarrier P, Jalil A,Vainchenker W, Louache F. Intracellular localization and constitutiveendocytosis of CXCR4 in human CD34� hematopoietic progenitor cells.Stem Cells. 2004;22:1015–1029.

10. Wang JF, Park IW, Groopman JE. Stromal cell-derived factor-1alphastimulates tyrosine phosphorylation of multiple focal adhesion proteinsand induces migration of hematopoietic progenitor cells: roles of phos-phoinositide-3 kinase and protein kinase C. Blood. 2000;95:2505–2513.

11. Ogaeri T, Eto K, Otsu M, Ema H, Nakauchi H. The actin polymerizationregulator WAVE2 is required for early bone marrow repopulation byhematopoietic stem cells. Stem Cells. 2009;27:1120–1129.

12. Basu S, Ray NT, Atkinson SJ, Broxmeyer HE. Protein phosphatase 2Aplays an important role in stromal cell-derived factor-1/CXC chemokineligand 12-mediated migration and adhesion of CD34� cells. J Immunol.2007;179:3075–3085.

13. Mazzone T, Chait A, Plutzky J. Cardiovascular disease risk in type 2diabetes mellitus: insights from mechanistic studies. Lancet. 2008;371:1800–1809.

14. McClung JA, Naseer N, Saleem M, Rossi GP, Weiss MB, Abraham NG,Kappas A. Circulating endothelial cells are elevated in patients with type2 diabetes mellitus independently of HbA(1)c. Diabetologia. 2005;48:345–350.

15. Vinik AI, Erbas T, Park TS, Nolan R, Pittenger GL. Platelet dysfunctionin type 2 diabetes. Diabetes Care. 2001;24:1476–1485.

16. Alberti KG, Zimmet P, Shaw J. Metabolic syndrome—a new world-widedefinition: a consensus statement from the International Diabetes Fed-eration. Diabet Med. 2006;23:469–480.

17. Ferroni P, Martini F, Riondino S, La FF, Magnapera A, Ciatti F,Guadagni F. Soluble P-selectin as a marker of in vivo platelet activation.Clin Chim Acta. 2009;399:88–91.

18. Dichiara J, Bliden KP, Tantry US, Hamed MS, Antonino MJ, Suarez TA,Bailon O, Singla A, Gurbel PA. The effect of aspirin dosing on plateletfunction in diabetic and nondiabetic patients: an analysis from the aspirin-induced platelet effect (ASPECT) study. Diabetes. 2007;56:3014–3019.

19. Piatkowski A, Groger A, Pantel M, Bozkurt A, Fuchs PC, Pallua N. Theextent of thermal injury affects fractions of mononuclear cells. Burns.2009;35:256–263.

20. Allan DS, Morgan SC, Birch PE, Yang L, Halpenny MJ, GunanayagamA, Li Y, Eapen L. Mobilization of circulating vascular progenitors incancer patients receiving external beam radiation in response to tissueinjury. Int J Radiat Oncol Biol Phys. 2009;75:220–224.

21. Rafat N, Hanusch C, Brinkkoetter PT, Schulte J, Brade J, Zijlstra JG, vander Woude FJ, van AK, Yard BA, Beck GC. Increased circulatingendothelial progenitor cells in septic patients: correlation with survival.Crit Care Med. 2007;35:1677–1684.

22. Yip HK, Chang LT, Chang WN, Lu CH, Liou CW, Lan MY, Liu JS,Youssef AA, Chang HW. Level and value of circulating endothelialprogenitor cells in patients after acute ischemic stroke. Stroke. 2008;39:69–74.

23. Massa M, Rosti V, Ferrario M, Campanelli R, Ramajoli I, Rosso R, DeFerrari GM, Ferlini M, Goffredo L, Bertoletti A, Klersy C, Pecci A,Moratti R, Tavazzi L. Increased circulating hematopoietic and endothelialprogenitor cells in the early phase of acute myocardial infarction. Blood.2005;105:199–206.

24. Heiss C, Amabile N, Lee AC, Real WM, Schick SF, Lao D, Wong ML,Jahn S, Angeli FS, Minasi P, Springer ML, Hammond SK, Glantz SA,Grossman W, Balmes JR, Yeghiazarians Y. Brief secondhand smokeexposure depresses endothelial progenitor cells activity and endothelialfunction: sustained vascular injury and blunted nitric oxide production.J Am Coll Cardiol. 2008;51:1760–1771.

25. Fox A, Smythe J, Fisher N, Tyler MP, McGrouther DA, Watt SM, HarrisAL. Mobilization of endothelial progenitor cells into the circulation inburned patients. Br J Surg. 2008;95:244–251.

26. Kozuka K, Kohriyama T, Nomura E, Ikeda J, Kajikawa H, Nakamura S.Endothelial markers and adhesion molecules in acute ischemic stroke:sequential change and differences in stroke subtype. Atherosclerosis.2002;161:161–168.

27. Fadini GP, de Kreutzenberg SV, Coracina A, Baesso I, Agostini C,Tiengo A, Avogaro A. Circulating CD34� cells, metabolic syndrome,and cardiovascular risk. Eur Heart J. 2006;27:2247–2255.

28. Fadini GP, Sartore S, Albiero M, Baesso I, Murphy E, Menegolo M,Grego F, Vigili de KS, Tiengo A, Agostini C, Avogaro A. Number and




ownloaded from


function of endothelial progenitor cells as a marker of severity fordiabetic vasculopathy. Arterioscler Thromb Vasc Biol. 2006;26:2140–2146.

29. Rossig L, Urbich C, Bruhl T, Dernbach E, Heeschen C, Chavakis E,Sasaki K, Aicher D, Diehl F, Seeger F, Potente M, Aicher A, Zanetta L,Dejana E, Zeiher AM, Dimmeler S. Histone deacetylase activity isessential for the expression of HoxA9 and for endothelial commitment ofprogenitor cells. J Exp Med. 2005;201:1825–1835.

30. Gampel A, Moss L, Jones MC, Brunton V, Norman JC, Mellor H. VEGFregulates the mobilization of VEGFR2/KDR from an intracellular endo-thelial storage compartment. Blood. 2006;108:2624–2631.

31. Sorkin A, von ZM. Endocytosis and signalling: intertwining molecularnetworks. Nat Rev Mol Cell Biol. 2009;10:609–622.

32. Zhang XF, Wang JF, Matczak E, Proper JA, Groopman JE. Janus kinase2 is involved in stromal cell-derived factor-1alpha-induced tyrosine phos-phorylation of focal adhesion proteins and migration of hematopoieticprogenitor cells. Blood. 2001;97:3342–3348.

33. De Falco E, Porcelli D, Torella AR, Straino S, Iachininoto MG, OrlandiA, Truffa S, Biglioli P, Napolitano M, Capogrossi MC, Pesce M. SDF-1

involvement in endothelial phenotype and ischemia-induced recruitmentof bone marrow progenitor cells. Blood. 2004;104:3472–3482.

34. Wijelath ES, Rahman S, Murray J, Patel Y, Savidge G, Sobel M.Fibronectin promotes VEGF-induced CD34 cell differentiation into en-dothelial cells. J Vasc Surg. 2004;39:655–660.

35. Hynes RO. Integrins: bidirectional, allosteric signaling machines. Cell.2002;110:673–687.

36. Stellos K, Bigalke B, Langer H, Geisler T, Schad A, Kogel A, Pfaff F,Stakos D, Seizer P, Muller I, Htun P, Lindemann S, Gawaz M. Expressionof stromal-cell-derived factor-1 on circulating platelets is increased inpatients with acute coronary syndrome and correlates with the number ofCD34� progenitor cells. Eur Heart J. 2009;30:584–593.

37. Petit I, Goichberg P, Spiegel A, Peled A, Brodie C, Seger R, Nagler A,Alon R, Lapidot T. Atypical PKC-zeta regulates SDF-1-mediatedmigration and development of human CD34� progenitor cells. J ClinInvest. 2005;115:168–176.

38. Sekiguchi H, Ii M, Losordo DW. The relative potency and safety ofendothelial progenitor cells and unselected mononuclear cells forrecovery from myocardial infarction and ischemia. J Cell Physiol. 2009;219:235–242.




ownloaded from


Tamsma, M.V. Huisman, A.J. Rabelink and A.J. van ZonneveldH.C. de Boer, M.M. Hovens, A.M. van Oeveren-Rietdijk, J.D. Snoep, E.J.P. de Koning, J.T.

Immobilization on Activated Platelets Cells After+ Cells Are Generated From Circulating CD34+/KDR+Human CD34

Print ISSN: 1079-5642. Online ISSN: 1524-4636 Copyright © 2010 American Heart Association, Inc. All rights reserved.

Greenville Avenue, Dallas, TX 75231is published by the American Heart Association, 7272Arteriosclerosis, Thrombosis, and Vascular Biology

doi: 10.1161/ATVBAHA.110.2168792011;31:408-415; originally published online October 28, 2010;Arterioscler Thromb Vasc Biol.

http://atvb.ahajournals.org/content/31/2/408World Wide Web at:

The online version of this article, along with updated information and services, is located on the

http://atvb.ahajournals.org/content/suppl/2010/10/28/ATVBAHA.110.216879.DC1Data Supplement (unedited) at:

http://atvb.ahajournals.org//subscriptions/

at: is onlineArteriosclerosis, Thrombosis, and Vascular Biology Information about subscribing to Subscriptions:

http://www.lww.com/reprints

Information about reprints can be found online at: Reprints:

document. Question and AnswerPermissions and Rightspage under Services. Further information about this process is available in the

which permission is being requested is located, click Request Permissions in the middle column of the WebCopyright Clearance Center, not the Editorial Office. Once the online version of the published article for

can be obtained via RightsLink, a service of theArteriosclerosis, Thrombosis, and Vascular Biologyin Requests for permissions to reproduce figures, tables, or portions of articles originally publishedPermissions:



ownloaded from

http://atvb.ahajournals.org/content/31/2/408

http://atvb.ahajournals.org/content/suppl/2010/10/28/ATVBAHA.110.216879.DC1

http://www.ahajournals.org/site/rights/

http://www.ahajournals.org/site/rights/

http://www.lww.com/reprints

http://atvb.ahajournals.org//subscriptions/


“Supplemental Material”

Methods

Isolation of CD34+ cells from cord blood

Umbilical cord blood (UCB) was collected immediately after delivery with written

approval in bags containing citrate, phosphate and dextrose (Maco Pharma,

Tourcoing, France). The mononuclear cell fraction was isolated by gradient

separation (Ficoll, Amersham, The Netherlands). CD34+ cell isolation was performed

using the CD34 progenitor cell isolation kit (Miltenyi, Biotech, The Netherlands).

Purity of the cells accounted for ≥ 95% as assessed by fluorescence activated cell

sorter (FACS) analysis (BD Biosciences), using FITC-labeled antibody against CD34.

CD34+ cells were kept at 4°C at a stock concentration of 107/mL in RPMI containing

1% human serum albumin (HSA, Sigma Aldrich, Germany) until they were used for

perfusion experiments.

Perfusions

Perfusions of CD34+ cells were performed as previously described.1 In short, a

confluent monolayer of EC was made procoagulant after 6 h stimulation with

recombinant tumor necrosis factor-α (10 ng/mL, Boehringer Ingelheim, Germany)

and the tissue factor-expressing EC were pre-perfused with platelet rich plasma

prepared from low molecular weight heparin anti-coagulated whole blood, leading to

the deposition of fibrin-rich platelet thrombi. This ex vivo perfusion model, mimicking

a vascular injury, was subsequently perfused with CD34+ cells, isolated from fresh

UCB at a shear rate of 1 dyne/cm2. Adhered CD34+ cells were then subjected to an

increased shear rate of 2 dyne/cm2 for 15 minutes. The shear-subjected, adherent

CD34+ cells were either fixed with 0.5% glutaraldehyde for 10 minutes or the

adherent cells were detached and recovered using flow-buffer containing 5 mmol/L

EDTA.

Cell characterization and enumeration by FACS

Freshly isolated CD34+ cells, not subjected to flow, or CD34+ cells either non-

attached during perfusion or attached and recovered after perfusion, were incubated

with blocking-buffer (phosphate-buffered saline (PBS) containing 3% bovine serum

albumin (BSA) and 2% fetal calf serum (FCS); 30 min at 4oC) and then incubated

with an isotype-matched phycoerythrin (PE)-, peridinin chlorophyll protein-cyanine

dye (PerCP-Cy5)- or fluorescein isothiocyanate (FITC)-labeled mouse

immunoglobulin (mIgG) or specific antibodies: PE-labeled mIgG directed against the

extracellular domain of KDR (R&D Systems, Abingdon, UK), Pacific Blue-labeled

mIgG against CD45 (Dakocytomation, Glostrup, Denmark), PerCP-Cy5-labeled mIgG

against CD34 (BD Biosciences, Breda, The Netherlands) or FITC-labeled mIgG

against PSGL-1 (KPL-1, St Cruz, CA, USA). FACS-analysis was performed using a

multi-parametric gating strategy based on the International Society of Hematotherapy

and Graft Engineering.2 During FACS-analysis, progenitor cells were characterized

as events located in the lymphocyte gate (based on forward- and side scatter

patterns) and expressing CD34pos and CD45dim. Of these gated cells KDR-, CD34-

and PSGL-1-expression was determined and expressed as MFI. Per experiment,

triplicate perfusions were performed and experiments were performed in duplicate.

In a separate experiment, freshly isolated CD34+ cells were fixed in 3.7%

paraformaldehyde (PFA) for 10 minutes at 4oC, washed in FACS-buffer (PBS

containing 1% BSA and 0.05% Na-azide) and half of these cells were permeabilized

using 0.5% Triton X-100 dissolved in FACS-buffer. These cells were incubated with

blocking-buffer and subsequently incubated with pure mIgG directed against the

extracellular domain of KDR (Reliatech), FITC-labeled mIgG against CD34 (BD

Biosciences) or isotype-matched controls. Binding of pure mIgG was detected with

Alexa 647-labeled anti-mIgG antibody (Molecular Probes, Leiden, The Netherlands).

During FACS-analysis, cells were selected in the lymphocyte gate and analyzed for

expression of CD34 and KDR (expressed as MFI). For the inhibition of PI3K, CD34+

cells were incubated with wortmannin (100 nM for 30 min at 37ºC) prior to perfusion.

Attached cells were harvested and stained for FACS-analysis parallel to non-treated

CD34+ cells. A detailed description of the FACS-based enumeration of circulating

CD34+ and CD34+/KDR+ cells has been published by van der Klaauw et al.3 For the

calculation of the number of CD34+ cells per 103 CD45pos-leukocytes, the events of

CD34+ cells was devided by the events of CD45+-leukocytes, multiplied by 1000.

Confocal Laser Scanning Microscopy

For the detection of surface-expressed KDR on shear-subjected CD34+ cells, glass

coverslips containing EC-adherent CD34+ cells were fixed with 0.5% glutaraldehyde,

washed with PBS, incubated with blocking-buffer and subsequently incubated with

mIgG against the extracellular domain of KDR (Reliatech). Binding of this primary

antibody was detected with Alexa 647-labeled anti-mIgG and fluorescent

micrographs were made, combined with Nomarski projection using CSLM (LSM 510,

Zeiss, Germany). For the detection of F-actin and p-Akt, glutaraldehyde-fixed

coverslips, containing EC-adherent CD34+ cells, were incubated with Triton to

permeabilize the cells (as described in the section “Cell characterization and

enumeration by FACS”), and subsequently incubated with blocking-buffer, phalloidin-

methylrhodamine (Sigma-Aldrich, Germany) and rabbit-IgG (rIgG) directed against p-

Akt (Cell Signalling, Bioké, The Netherlands). Binding of the primary antibody was

detected with Alexa-647 labeled anti-rIgG (Molecular Probes, Leiden, The

Netherlands). For staining of early-endosome-associated proteins, freshly isolated

CD34+ cells were fixed in 3.7% PFA and either or not permeabilized with Triton. Cells

were blocked and incubated with mIgG against KDR (Reliatech) or mIgG against the

CXC chemokine receptor-4 (CXCR4; R&D Systems), both in combination with rIgG

against early endosome antigen-1 (EEA-1, kindly provided by Dr. S. Urbé,

Physiological Laboratory, University of Liverpool, UK). Binding of primary antibodies

was detected with secondary antibodies against mIgG labeled with Alexa-647 and

anti-rIgG labeled with Alexa-568 (both from Molecular Probes). Fluorescent images

were combined with Nomarski projection using CLSM.

Scanning electron microscopy

After perfusion, samples were dehydrated, critical-point dried (Bal-Tec CPD 030,

Liechtenstein) and coated with gold/palladium using a sputtercoater (EMI-TEC

K500X, Berlin, Germany). Samples were evaluated at 5 kV using a JEOL-JSM-6700F

scanning electron microscope (Jeol Ltd, Japan).

Subjects and study design

Circulating cells were studied in different samples: human BM, UCB, PB of G-CSF-

stimulated healthy donors, PB of diabetes mellitus type 2 (DM) patients and PB of

age-matched control subjects. BM and UCB was obtained from subjects with

informed consent. G-CSF mobilization was performed in healthy subjects, prepared

to donate stem cells for allogeneic peripheral blood stem cell transplantation. Stem

cell donors entered the study according to a protocol approved by the institutional

medical ethics committee. Stem cell donors received a single dose of 10 µg/kg/day

recombinant human G-CSF (Filgrastim; Amgen, Thousand Oaks, CA) one and five

days before large-volume leukapheresis. PB samples were assayed for numbers of

CD34+- and CD34+/KDR+-cells one day before leukapheresis.

Subjects with DM were recruited from general practitioners affiliated to the Leiden

University Medical Center (The Netherlands). Details about the randomized placebo-

controlled study on the effect of aspirin in subjects with type 2 diabetes have been

published previously.4 All subjects gave written informed consent and the study was

approved by the institutional review committee and performed in accordance with the

Declaration of Helsinki. The aspirin-study had a prospective, randomized, placebo-

controlled, double-blind, crossover design. All subjects (n=20) received one period

placebo and the other period aspirin (300 mg/day). The first treatment period with

aspirin or placebo for 6 weeks was followed by a washout period of 4 weeks.

Thereafter, those assigned to placebo in the first period received aspirin for 6 weeks

and those assigned to aspirin received placebo for additional 6 weeks. At each visit,

EDTA-anticoagulated PB samples were drawn from antecubital veins.

The age-matched controls comprised male subjects above 50 years of age with mild

visceral obesity, but without metabolic syndrome (MetS),5 type 2 diabetes mellitus,

manifest cardiovascular disease, use of statins or non-steroidal anti-inflammatory

drugs, current smoking, familial history of premature cardiovascular disease and

severe obesity (BMI>40 kg/m²). All control subjects gave informed consent and the

study complied with the Declaration of Helsinki.

Measurement of sP-sel and VEGF.

sP-Sel and VEGF was measured in plasma using commercially available ELISA kits

(both R&D Systems), according to the instructions of the manufacturers.

Measurements were performed in duplicate and average values were used.

Statistical Analysis

Data are expressed as mean ± standard error of the mean (SEM). Statistical one-way

analysis was performed by analysis of variance (ANOVA), unpaired t-test analysis

when the distribution was parametric or Mann-Whitney analysis when the distribution

was non-parametric, using the computer program GraphPad Prism 4.0 for Windows.

A value of P<0.05 indicated statistical significance.

Effects of aspirin versus placebo on circulating cells were estimated using paired

samples t-tests, since the differences in circulating progenitor cells between

treatment with aspirin and placebo were normally distributed (data not shown).

Correlations with delta sP-sel and delta VEGF were calculated using linear

regression analysis.

Reference List

(1) de Boer HC, Verseyden C, Ulfman LH, Zwaginga JJ, Bot I, Biessen EA, Rabelink TJ, van Zonneveld AJ. Fibrin and activated platelets cooperatively guide stem cells to a vascular injury and promote differentiation towards an endothelial cell phenotype. Arterioscler Thromb Vasc Biol 2006 July;26(7):1653-9.

(2) Keeney M, Chin-Yee I, Weir K, Popma J, Nayar R, Sutherland DR. Single platform flow cytometric absolute CD34+ cell counts based on the ISHAGE guidelines. International Society of Hematotherapy and Graft Engineering. Cytometry 1998 April 15;34(2):61-70.

(3) van der Klaauw AA, Pereira AM, Rabelink TJ, Corssmit EP, Zonneveld AJ, Pijl H, de Boer HC, Smit JW, Romijn JA, de Koning EJ. Recombinant human GH replacement increases CD34+ cells and improves endothelial function in adults with GH deficiency. Eur J Endocrinol 2008 August;159(2):105-11.

(4) Hovens MM, Snoep JD, Groeneveld Y, Frolich M, Tamsma JT, Huisman MV. Effects of aspirin on serum C-reactive protein and interleukin-6 levels in patients with type 2 diabetes without cardiovascular disease: a randomized placebo-controlled crossover trial. Diabetes Obes Metab 2008 August;10(8):668-74.

(5) Alberti KG, Zimmet P, Shaw J. Metabolic syndrome--a new world-wide definition. A Consensus Statement from the International Diabetes Federation. Diabet Med 2006 May;23(5):469-80.

B

H

DF

KDREEA-1

CXCR4

KDRCXCR4

*

*

EEA-1

KDREEA-1

EEA-1CXCR4

*rIgG

mIgG

rIgG

A

B

C

D

E

G mIgG

Supplemental Figure I

Supplemental Figure I. KDR expressed by CD34+ cells is associated with an early

endosomal compartment. To study co-localization of KDR with an early endosome

compartment, CD34+ cells were fixed and either permeabilized (no asterix) or non-

permeabilized (asterix). Cells were stained for (A and D) KDR (green) and EEA-1

(red), for (B and E) CXCR4 (green) and EEA-1 (red) or for (C) CXCR4 (green) and

KDR (red). Background levels were visualized with (F and G) mIgG (green) and rIgG

(red). Fluorescent imaging was combined with Nomarski projections, leading to co-

localizing signals in yellow. Scale bars represent 5 µm.

Documents

Human CD34 /KDR Cells Are Generated From Circulating CD34