6
825 J. Endocrinol. Invest. 36: 825-830, 2013 DOI: 10.3275/9020 ABSTRACT. Background: Endothelial progenitor cells (EPCs), involved in the repairing mechanisms of vascular damage, are positively correlated to insulin-like growth factor I (IGF-I) con- centrations in healthy adults. However, the levels of EPCs and their role in acromegalic patients have never been investi- gated. Aim: We conducted a cross-sectional study in order to assess the levels of the different phenotypes of circulating EPC in acromegalic patients. Subjects and methods: The study was performed at the Endocrinology Unit of Federico II Uni- versity and at the Unit of Metabolic Diseases and Endocrino- logy of the Second University of Naples. Fifty-five acrome- galic patients and 65 healthy controls were studied. EPCs were assessed by flow cytometry and IGF-I by immunoradio- metric assay. Results: Compared with subjects of the control group, acromegalic patients showed significantly higher lev- els of EPCs phenotypes expressing KDR antigen [KDR+, cells per 10 6 events, median and interquartile range, 44 (28-67) vs 23 (13-40), p=0.006; CD34+KDR+ 25 (18-38) vs 12 (8-17), p<0.001; CD133+KDR+ 17 (13-30) vs 8 (6-12), p<0.001; CD34+KDR+CD133+ 16 (12-25) vs 8 (6-10), p<0.001]. There was a positive correlations between CD34+KDR+CD133+ cells count and IGF-I in acromegaly group (r=0.79, p<0.001). Conclusions: Acromegalic patients show higher circulating EPCs levels expressing KDR, positively correlated with IGF-I, suggesting a role for IGF-I in regulating the expression of this surface marker in the early phase of EPCs differentiation. (J. Endocrinol. Invest. 36: 825-830, 2013) © 2013, Editrice Kurtis INTRODUCTION Serum insulin-like growth factor I (IGF-I) levels in top quar- tile of the normal range are associated with reduced vas- cular tone, increased survival of endothelial and smooth muscle cells, and reduced circulating levels of inflamma- tory cytokines (1). On the other hand, GH deficiency with low IGF-I levels is associated with reduced nitric oxide (NO) release by endothelial cells (2), increased carotid intima-media thickness and ischemic events (3, 4). Acromegaly is a rare syndrome characterized by an ex- cess of GH and IGF-I, most commonly due to a GH-pro- ducing pituitary adenoma. Cardiomyopathy, manifested as concentric left ventricular hypertrophy and diastolic and systolic dysfunction, is one of the most important complications of acromegaly, responsible for cardiovas- cular (CV) death. Although some research found evidence for endothelial dysfunction in acromegaly (5, 6), there is much more evidence that the extent of atherosclerotic involvement is not different from control people (7, 8). Circulating endothelial progenitor cells (EPCs) represent bone-marrow-derived stem cells able to differentiate in mature endothelial cells, involved in the repairing mech- anisms of vascular damage. They are recently emerging as a novel surrogate risk factor of CV diseases (9-11). In healthy men, the number of EPCs correlates positively with endothelial function and negatively with the Fram- ingham CV risk score (9). Moreover, reduced levels and functional impairment of EPCs, which correlated with CV risk factors, have been reported in patients with coronary artery diseases (CAD) (10). Lastly, in 519 patients with CAD, increased levels of EPCs were associated with a 69% reduced risk of death from CV causes (11). Recently, it has been noticed a positive correlation be- tween GH-mediated increase in serum IGF-I and circu- lating EPCs in healthy adults (12, 13). Thum et al. (14) demonstrated that in elderly people the number of EPC was reduced and correlated with a reduction of IGF-I lev- els; replacement therapy with GH or IGF-I improved EPC number and function. In theory, clinical disorders char- acterized by high circulating GH and IGF-I levels may be associated with increased EPCs levels, but this has nev- er been demonstrated. The aim of the present study was to assess: a) the number of circulating EPCs in acrome- galic patients and in healthy controls and b) the correla- tion between EPCs count and IGF-I levels. MATERIALS AND METHODS Participants Fifty-five patients affected by acromegaly and 65 non-acrome- galic subjects sex-, age- and body mass index (BMI)-matched with the patients were enrolled in the study. All the patients had to have a diagnosis of acromegaly according to established cri- teria (15). Patients were recruited among those attending both the Endocrinology Unit at the “Federico II” University of Naples, and the Unit of Endocrinology and of Metabolic Diseases at the Second University of Naples, Italy. Among these patients, 16 were newly diagnosed with active acromegaly, while 39 had a mean disease duration of 5 yr. At the time of the enrollment, 20 patients were untreated (16 newly diagnosed, 4 with normal lev- els of IGF-I after pituitary surgery), and the other 35 were on Key-words: Acromegaly, cardiovascular risk, endothelial dysfunction, EPCs, IGF-I. Correspondence: K. Esposito, MD, PhD, Endocrinology and Metabolic Diseases Unit, Department of Cardio-Thoracic and Respiratory Sciences, Second University of Naples, Piazza L. Miraglia 2, 80138 Napoli, Italy. E-mail: [email protected] Accepted April 3, 2013. First published online June 26, 2013. Circulating endothelial progenitor cells in acromegaly G. Bellastella 1 , M.I. Maiorino 1 , R. Pivonello 2 , L.F.S. Grasso 2 , M. Galdiero 2 , A.A. Sinisi 3 , A. Colao 2 , D. Giugliano 1 , and K. Esposito 3 1 Unit of Metabolic Diseases, Department of Geriatrics and Metabolic Diseases, Second University of Naples; 2 Unit of Endocrinology, Department of Clinical and Molecular Endocrinology and Oncology, Federico II University of Naples; 3 Unit of Endocrinology, Department of Cardio-Thoracic and Respiratory Sciences, Second University of Naples, Naples, Italy

Circulating endothelial progenitor cells in acromegaly

Embed Size (px)

Citation preview

Page 1: Circulating endothelial progenitor cells in acromegaly

825

J. Endocrinol. Invest. 36: 825-830, 2013DOI: 10.3275/9020

ABSTRACT. Background: Endothelial progenitor cells (EPCs),involved in the repairing mechanisms of vascular damage, arepositively correlated to insulin-like growth factor I (IGF-I) con-centrations in healthy adults. However, the levels of EPCs andtheir role in acromegalic patients have never been investi-gated. Aim: We conducted a cross-sectional study in order toassess the levels of the different phenotypes of circulatingEPC in acromegalic patients. Subjects and methods: The studywas performed at the Endocrinology Unit of Federico II Uni-versity and at the Unit of Metabolic Diseases and Endocrino-logy of the Second University of Naples. Fifty-five acrome-galic patients and 65 healthy controls were studied. EPCswere assessed by flow cytometry and IGF-I by immunoradio-metric assay. Results: Compared with subjects of the control

group, acromegalic patients showed significantly higher lev-els of EPCs phenotypes expressing KDR antigen [KDR+, cellsper 106 events, median and interquartile range, 44 (28-67) vs23 (13-40), p=0.006; CD34+KDR+ 25 (18-38) vs 12 (8-17),p<0.001; CD133+KDR+ 17 (13-30) vs 8 (6-12), p<0.001;CD34+KDR+CD133+ 16 (12-25) vs 8 (6-10), p<0.001]. Therewas a positive correlations between CD34+KDR+CD133+cells count and IGF-I in acromegaly group (r=0.79, p<0.001).Conclusions: Acromegalic patients show higher circulatingEPCs levels expressing KDR, positively correlated with IGF-I,suggesting a role for IGF-I in regulating the expression of thissurface marker in the early phase of EPCs differentiation.(J. Endocrinol. Invest. 36: 825-830, 2013)©2013, Editrice Kurtis

INTRODUCTION

Serum insulin-like growth factor I (IGF-I) levels in top quar-tile of the normal range are associated with reduced vas-cular tone, increased survival of endothelial and smoothmuscle cells, and reduced circulating levels of inflamma-tory cytokines (1). On the other hand, GH deficiency withlow IGF-I levels is associated with reduced nitric oxide(NO) release by endothelial cells (2), increased carotidintima-media thickness and ischemic events (3, 4).Acromegaly is a rare syndrome characterized by an ex-cess of GH and IGF-I, most commonly due to a GH-pro-ducing pituitary adenoma. Cardiomyopathy, manifestedas concentric left ventricular hypertrophy and diastolicand systolic dysfunction, is one of the most importantcomplications of acromegaly, responsible for cardiovas-cular (CV) death. Although some research found evidencefor endothelial dysfunction in acromegaly (5, 6), there ismuch more evidence that the extent of atheroscleroticinvolvement is not different from control people (7, 8).Circulating endothelial progenitor cells (EPCs) representbone-marrow-derived stem cells able to differentiate inmature endothelial cells, involved in the repairing mech-anisms of vascular damage. They are recently emergingas a novel surrogate risk factor of CV diseases (9-11). Inhealthy men, the number of EPCs correlates positivelywith endothelial function and negatively with the Fram-

ingham CV risk score (9). Moreover, reduced levels andfunctional impairment of EPCs, which correlated with CVrisk factors, have been reported in patients with coronaryartery diseases (CAD) (10). Lastly, in 519 patients withCAD, increased levels of EPCs were associated with a69% reduced risk of death from CV causes (11).Recently, it has been noticed a positive correlation be-tween GH-mediated increase in serum IGF-I and circu-lating EPCs in healthy adults (12, 13). Thum et al. (14)demonstrated that in elderly people the number of EPCwas reduced and correlated with a reduction of IGF-I lev-els; replacement therapy with GH or IGF-I improved EPCnumber and function. In theory, clinical disorders char-acterized by high circulating GH and IGF-I levels may beassociated with increased EPCs levels, but this has nev-er been demonstrated. The aim of the present study wasto assess: a) the number of circulating EPCs in acrome-galic patients and in healthy controls and b) the correla-tion between EPCs count and IGF-I levels.

MATERIALS AND METHODSParticipantsFifty-five patients affected by acromegaly and 65 non-acrome-galic subjects sex-, age- and body mass index (BMI)-matchedwith the patients were enrolled in the study. All the patients hadto have a diagnosis of acromegaly according to established cri-teria (15). Patients were recruited among those attending boththe Endocrinology Unit at the “Federico II” University of Naples,and the Unit of Endocrinology and of Metabolic Diseases at theSecond University of Naples, Italy. Among these patients, 16were newly diagnosed with active acromegaly, while 39 had amean disease duration of 5 yr. At the time of the enrollment, 20patients were untreated (16 newly diagnosed, 4 with normal lev-els of IGF-I after pituitary surgery), and the other 35 were on

Key-words: Acromegaly, cardiovascular risk, endothelial dysfunction, EPCs, IGF-I.

Correspondence: K. Esposito, MD, PhD, Endocrinology and Metabolic Diseases Unit,Department of Cardio-Thoracic and Respiratory Sciences, Second University of Naples,Piazza L. Miraglia 2, 80138 Napoli, Italy.

E-mail: [email protected]

Accepted April 3, 2013.

First published online June 26, 2013.

Circulating endothelial progenitor cells in acromegalyG. Bellastella1, M.I. Maiorino1, R. Pivonello2, L.F.S. Grasso2, M. Galdiero2, A.A. Sinisi3, A. Colao2,D. Giugliano1, and K. Esposito3

1Unit of Metabolic Diseases, Department of Geriatrics and Metabolic Diseases, Second University of Naples;2Unit of Endocrinology, Department of Clinical and Molecular Endocrinology and Oncology, Federico II University of Naples;3Unit of Endocrinology, Department of Cardio-Thoracic and Respiratory Sciences, Second University of Naples, Naples, Italy

Page 2: Circulating endothelial progenitor cells in acromegaly

G. Bellastella, M.I. Maiorino, R. Pivonello, et al.

826

pharmacological therapy with somatostatin analogues (SSA).Twenty-two patients had already undergone transphenoidalsurgery. The controls were recruited from the medical andparamedical staff, and served as internal reference. They had tohave normal IGF-I levels for age, normal renal and liver function,no clinical evidence of diabetes mellitus, peripheral vascular dis-ease or coronary artery disease, no history of smoking, and nocurrent use of drugs able to increase EPCs [statins, angiotensin-converting enzyme (ACE) inhibitors, and glitazones]. All subjectsgave informed consent to participate in the study that was ap-proved by the local Institutional Review Board.

Quantification of circulating EPCsCirculating EPCs are characterized by the expression of imma-ture markers, such as CD34 and CD133, and endothelial markers,such as KDR. With the use of these surface markers, seven sub-populations displaying different antigenic profiles could be iden-tified: CD34+, CD133+, KDR+, CD34+CD133+, CD34+KDR+,CD133+KDR+, and CD34+ KDR+CD133+.Fasting blood samples were processed within 1-2 h. Analysiswas performed on 500 μl of peripheral blood incubated with flu-orescein isothiocyanate (FITC)-conjugated antihuman CD34monoclonal antibody (mAb) (Becton Dickinson, Buccinasco, Mi-lano, Italy), phycoerythrin (PE)-conjugated antihuman KDR mAb(Becton Dickinson, Buccinasco, Milano, Italy), and allophyco-cyanin (APC)-conjugated antihuman CD133 (Miltenyi Biotec,Calderara di Reno, Bologna, Italy). Then, cells were lysed in alysing buffer, washed twice with PBS 1% FCS and then analysedby flow cytometry. Isotope immunoglobulin IgG1 and IgG2a an-tibody was used to discriminate between signal range and base-line fluorescence within the samples. Peripheral blood cells wereanalyzed for the expression of surface antigens by direct flow

cytometry, as previously described (16, 17). In brief, quantita-tive analysis was performed on a BD FACSCalibur cytometer(Becton Dickinson), and 1,000,000 cells were acquired in eachsample. A morphological gate was used to exclude granulo-cytes. Then, we gated CD34+ or CD133+ peripheral blood cellsin the mononuclear cell fraction and examined the resulting pop-ulation for the dual expression of KDR. In the two-dimensionaldotplot analysis, we identified CD34+CD133+ cells. Total KDR+mononuclear cells were identified separately. Triple-positivecells were identified by the dual expression of KDR and CD133in the CD34+ gate (Fig. 1). Data were processed with the use ofthe Macintosh CELLQuest software program (Becton Dickinson).Measures were repeated twice in two separate blood samples.The instrument setup was optimized daily by analyzing the ex-pression of peripheral blood lymphocytes labeled with an anti-CD4 FITC/CD8 PE/CD3 PECy5/CD45APC 4-color combination.The same trained operator, who was blinded to the subjects’characteristics, performed all of the tests throughout the study.

Anthropometric measures and laboratory analysesHeight and weight were recorded, with the participants wear-ing lightweight clothing and no shoes, using a Seca 200 scale(Seca, Hamburg, Germany) with attached stadiometer. BMI wascalculated as weight in kg divided by the square of height inmeters (kg/m2). Blood pressure (BP) was also measured in a qui-et environment with a mercury sphygmomanometer with thepatient in a sitting position after 5min of rest. Systolic and dias-tolic BP values represented the mean of three different read-ings measured at 5min intervals. A standard electrocardiographyregistration, echocardiography and 24-h BP monitoring wereperformed in all patients.Waist circumference was measured at a level midway between

100 101 102 103 104

CD133 APC

KD

RP

E10

010

110

210

310

4

A

100 101 102 103 104

CD133 APC

KD

RP

E10

010

110

210

310

4

B

100 101 102 103 104

CD133 APC

KD

RP

E10

010

110

210

310

4

C

100 101 102 103 104

CD133 APC

KD

RP

E10

010

110

210

310

4

D

Fig. 1 - Representative flow cytometricdot plots showing high (A), intermediate(B) and low (C) number of CD34+KDR+CD133+ in acromegalic patients and in acontrol subject (D). Triple-positive cellswere identified by the dual expression ofKDR and CD133 in the CD34+ gate (up-per-right quadrant).

Page 3: Circulating endothelial progenitor cells in acromegaly

EPCs in acromegaly

827

the lowest rib and the iliac crest. GH and IGF-I values were mea-sured by radioimmunoassay using commercial kits (IGF-I-RIACT,Cisbio Bioassays, Bagnols sur Ceze, France). The IGF-I referenceranges in our laboratory for the relevant ages were: 103-351ng/ml (29-40 yr); 105-328 ng/ml (41-50 yr); 99-295 ng/ml (51-60yr); 91-222 ng/ml (61-70 yr). Assays for glucose, total, low-den-sity, and high-density lipoprotein cholesterol, and triglyceridelevels were performed in the hospital’s chemistry laboratory.

Statistical analysisThe primary outcome was the level of circulating EPCsCD34+KDR+CD133+, which is recognized as the putative EPCphenotype (18, 19). We calculated the sample size on the pri-mary outcome: assuming an α level of 0.05 and 90% power, therequired number of patients for each group to observe aCD34+KDR+CD133+ difference of 5 (standard deviation 5) is23. Data are reported as mean±SD or median and interquartilerange. The results from flow cytometry are expressed as thenumber of cells per 106 events. Differences between two ormore groups were evaluated by rank sum test or two-sided Stu-dent’s t-test, depending on the normality of the sample. Statis-tical associations between variables were assessed using Spear-man’s rank correlation test and multivariate regression analysis.Statistical significance was assumed at a p level of <0.05. Allanalyses were performed using Sigmastat 3_5 web.

RESULTS

The clinical and metabolic characteristics of the partici-pants in the study are summarized in Table 1. The an-thropometric parameters, BMI and waist circumference,

the blood lipid and BP levels were not significantly dif-ferent between the two groups. All patients showed nor-mal kidney and liver function. Thirty-eight patients (69%)had hypertension, 33 (60%) showed a left ventricular hy-pertrophy (LVH) and nobody had history of previous CVevents. Among the patients on somatostatin analoguestreatment, 13 had concomitant therapy including met-formin, statins, beta blockers, calcium antagonists, andACE-inhibitors. GH, IGF-I, fasting blood glucose, and gly-cated hemoglobin (HbA1c) levels were significantly higherin acromegalic patients than in the control group.Acromegalic patients showed significantly higher levelsof EPCs phenotypes expressing KDR antigen than con-trols, namely KDR+, CD34+KDR+, CD133+KDR+,CD34+KDR+CD133+ (Table 2). No significant differencesin the count of the other phenotypes (CD34+, CD133+,CD34+CD133+) were recorded. EPCs count was not sig-

Cases Controls pNo.=55 No.=65

Age (yr) 57.6±10.8 53±10 0.531

Sex (M/F) 22/33 30/35 /

Duration of disease (yr) 5±3.9 / /

Body mass index (kg/m2) 30.4±5.2 28.3±4.9 0.247

Waist (cm) 98±14 92±12 0.423

Hypertension, no. (%) 38 (69%) /

Left ventricular hypertrophy 33 (60%) /

Glucose 104 (96-130) 89 (85-92) 0.001

HbA1c 6.06±0.4 5.3±0.4 <0.001

HOMA of insulin sensitivity 6.8±5.1 2.1±0.8 <0.001

Lipids (mg/dl)Cholesterol 206±48 199±30 0.660High density lipoprotein cholesterol 52±19 58±12 0.121Low density lipoprotein cholesterol 129±42 122±24 0.590Triglyceride 100±34 125±31 0.051

HormonesGH (ng/ml) 2.9 (2.4-3.7) 1.1 (0.1-1.8) <0.001IGF-I (ng/ml) 425 (239-580) 174 (154-195) <0.001

TherapySomatostatin analogs 35/55 0 /Metformin 6Ace-inhibitors 2 0 /Beta blockers+Calcium antagonists 2 0 /Statin+Metformin 2 0 /Metformin+Ace-inhibitors+Statin 1 0 /

Values are expressed as mean±SD, percentage, or median (interquartile range). HbA1c: glycated hemoglobin; HOMA: homeostasis model assessment.

Table 1 - Clinical characteristics of the study subjects.

EPC (n/106 events) Cases Controls p

CD34+ 262 (198-336) 217 (157-293) 0.15

CD34+CD133+ 101 (73-168) 106 (82-130) 1.0

CD133+ 145 (115-226) 142 (104-177) 0.43

KDR+ 44 (28-67) 23 (13-40) 0.006

CD34+KDR+ 25 (18-38) 12 (8-17) <0.001

CD133+KDR+ 17 (13-30) 8 (6-12) <0.001

CD34+KDR+133+ 16 (12-25) 8 (6-10) <0.001

EPC: endothelial progenitor cells.

Table 2 - Circulating levels of seven progenitor cell phenotypes.

Page 4: Circulating endothelial progenitor cells in acromegaly

G. Bellastella, M.I. Maiorino, R. Pivonello, et al.

828

nificantly different between sexes, both in acromegalic pa-tients and in the control group. We did not find any signif-icant difference in CD34+KDR+CD133+ cells count be-tween treated and untreated patients (Fig. 2). Moreover,acromegalic patients taking SSA analogues and con-comitant therapy (no.=13) did not show any significantdifferences in EPCs count compared with those treatedonly with SSA (no.=22) or untreated (no.=20) (Fig. 2).We also evaluated the correlation between IGF-I levelsand EPCs phenotypes count. There was a significantlypositive correlation between the KDR+ phenotype andIGF-I levels in control subjects (Table 3), and a signifi-cantly positive correlation between circulatingCD34+KDR+CD133+ cells and IGF-I levels in acromegaly(Table 3, Fig. 3). In the multivariate regression analysis,with EPCs as the dependent variable, and HbA1c, glu-cose, GH, and IGF-I as independent variables, only IGF-I was significantly associated with the dependent vari-able (β-coefficient: 0.0365, p<0.001).

DISCUSSION

We report the first assessment of different EPCs pheno-types in acromegaly. We found that acromegalic patientshad significantly higher levels of circulating KDR+,CD34+KDR+, CD133+KDR+ and CD34+KDR+CD133+cells than control subjects, with a positive correlation be-tween the number of circulating CD34+KDR+CD133+cells and the IGF-I levels.All EPCs phenotypes presenting the KDR antigen, alsoknown as vascular endothelial growth factor receptortype 2 (VEGFR2), were significantly increased in acrome-galic patients compared with controls. This receptor,known as kinase insert domain receptor, is a type III re-ceptor tyrosine kinase. It functions as the main mediatorof VEGF-induced endothelial proliferation, survival, mi-gration, tubular morphogenesis and sprouting (20).Hence, it is considered necessary for neo-angiogeneticprocesses. Piecewics et al. (21), reported an increasedexpression of the VEGFR2 mRNA when embryonal stemcells were exposed to IGF-I.The elevated circulating levels of EPCs observed in ourpatients may be the results of the IGF-I mediated NO in-crease. IGF-I plays an important role in maintaining bio-logical function of endothelium, stimulating NO produc-tion by endothelial cells, regulating the vascular tone andexerting many others anti-atherogenic properties (22).On the other hand, NO acts as an important regulatingfactor of EPCs, both stimulating the release of EPC frombone marrow and promoting their mobilization towardsthe sites of vascular damage (14, 23, 24).As EPCs show a wide heterogeneity with functionally im-portant subpopulations, there is no clear consensus onwhich antigenic profile best identifies progenitor cells withthe potential to repair the endothelium. Three main anti-genic profiles have been proposed: CD34 is an adhesionmolecule expressed both on hematopoietic stem cells and

EPC phenotypes Acromegaly Controlsno.=55 no.=65

r p r p

CD34+ –0.11 0.444 –0.32 0.061

KDR+ 0.04 0.852 0.40 0.020

CD133+ –0.04 0.803 –0.13 0.511

CD34+CD133+ –0.13 0.401 –0.01 0.865

CD34+KDR+ 0.11 0.321 –0.15 0.442

CD133+KDR+ 0.15 0.512 0.12 0.562

CD34+KDR+CD133+ 0.79 <0.001 0.18 0.313

EPC: Endothelial progenitor cells.

Table 3 - Correlation between EPC phenotypes and IGF-I levelsin acromegaly and control group.

CD

34+

KD

R+

CD

133+

30

20

10

0No Treatment Treatment

treatment (SSA) (SSA+CT)

Fig. 2 - Subgroup analysis: CD34+KDR+CD133+ cells count dif-ferences between untreated (no.=20) and treated patients (SSAno.=25; SSA+CT no.=13). SSA: somatostatin analogues; CT:concomitant therapy.

IGF-

I(ng

/ml)

900

850

800

750

700

650

600

550

500

450

400

350

300

250

200

150

0

CD34+KDR+CD133+Cell Count

p<0.001 r=0.79

0 10 20 30 40

Fig. 3 - Correlation between CD34+KDR+CD133+ cells countand IGF-I in acromegalic patients; those below the horizontalline are well controlled.

Page 5: Circulating endothelial progenitor cells in acromegaly

EPCs in acromegaly

829

other progenitor cells, as well as on activated endothelium,and is typically considered a marker of immaturity; CD133is a surface antigen of unknown function that identifies moreimmature progenitor cells than CD34 alone, characterizinghematopoietic cell lines; KDR represents the VEGFR andindicates early endothelial differentiation (25, 26).The results of previous studies indicate that theCD34+KDR+ cells levels independently predict CV eventsand atherosclerosis progression in patients with CAD (11,27) and negatively correlate with values of carotid intima-media thickness (28). CD34+/KDR+/CD133+ cells may beidentified as a more immature cellular subset than thoseexpressing CD34 and KDR, but not CD133 (18, 19). Gen-erally, the detection of CD34+KDR+CD133+ cells in theperipheral blood may be difficult due to their low num-ber (26) and their rapid differentiation to more maturephenotype. However, we found that circulating levels ofthis phenotype are significantly increased in acromegalicpatients, with a positive correlation with IGF-I. The de-tection of high levels of CD34+/KDR+/CD133+ cells inacromegalic patients, independently of disease duration,suggests an elevated turnover of this cellular subset, like-ly in response to high levels of IGF-I.The role of IGF-I on the CV system is still debated. In thegeneral population, IGF-I levels are inversely correlatedwith ischemic heart disease, CV death, and stroke in thegeneral population (6, 29-31); moreover, the impairmentof flow mediated dilation correlates with low IGF-I levels(22). In acromegaly, CAD risk is low and remains stable af-ter successful treatment, leading to suggest that GH ex-cess per se does not carry an additional CAD risk (32).Acromegalic patients do not show an increased carotidintima-media thickness compared with non acromegalicmatched control subjects (7, 8). It seems reasonable tospeculate that the greater EPCs count expressing theKDR phenotype found in our patients may be involvedin the less than expected atherosclerotic burden ofacromegaly. Supporting this working hypothesis, GH-me-diated increase in IGF-I levels is associated with im-provement of EPCs number in healthy adults (13) and inelderly people (12). A two-fold increase of EPCs, in par-ticular of KDR expressing phenotype, has been reportedin middle aged healthy men after GH treatment (14).This study has limitations. The cross-sectional nature ofthe study does not allow to firmly establish a cause andeffect relationship between variables. On the other hand,we analysed 55 patients, which seems a suitable numberfor a rare disease, with an adequate statistical power.Thirteen out of 55 patients reported the use of con-comitant therapy, but they did not differ in terms of IGF-I and EPC levels from other acromegalic patients takingonly SSA. Moreover, although all acromegalic patientspresented clinical factors potentially influencing in a neg-ative way the circulating EPCs levels, including abnormalglucose metabolism, obesity and hypertension (33-35),all cases showed a higher EPCs number than controls,suggesting a potential role of IGF-I in neutralizing thenegative effect played by these conditions.In conclusion, acromegalic patients show higher circu-lating EPCs levels expressing KDR than healthy controls,with a positive correlation between IGF-I levels andCD34+KDR+CD133+ cells count, suggesting a direct

role of IGF-I in regulating the expression of this surfacemarker in the early phase of EPCs differentiation. Al-though acromegaly is characterized by several metabol-ic and vascular disorders, our results suggest that EPCsmay be involved in the lower than expected atheroscle-rotic risk of acromegalic patients.

ACKNOWLEDGMENTSDeclaration of interestThe authors declare that there is no conflict of interest that could be per-ceived as prejudicing the impartiality of the research reported.

REFERENCES1. Colao A. The GH-IGF-I axis and the cardiovascular system: clinical

implications. Clin Endocrinol (Oxf) 2008, 69: 347-58.2. Thum T, Fleissner F, Klink I, et al. Growth hormone treatment im-

proves markers of systemic nitric oxide bioavailability via insulin-like growth factor-I. J Clin Endocrinol Metab 2007, 92: 4172-9.

3. Colao A, Di Somma C, Filippella M, et al. Insulin-like growth factor-1 deficiency determines increased intima-media thickness at com-mon carotid arteries in adult patients with growth hormone defi-ciency. Clin Endocrinol (Oxf) 2004, 61: 360-6.

4. Juul A, Scheike T, Davidsen M, Gyllenborg J, Jørgensen T. Lowserum insulin-like growth factor I is associated with increased riskof ischemic heart disease: a population-based case-control study.Circulation 2002, 106: 939-44.

5. Akgul E, Tokgozoglu SL, Erbas T, et al. Evaluation of the impactof treatment on endothelial function and cardiac performance inacromegaly. Echocardiography 2010, 27: 990-6.

6. Kartal I, Oflaz H, Pamukçu B, et al. Investigation of early atheroscle-rotic changes in acromegalic patients. Int J Clin Pract 2010, 64: 39-44.

7. Otsuki M, Kasayama S, Yamamoto H, et al. Characterization of pre-mature atherosclerosis of carotid arteries in acromegalic patients.Clin Endocrinol (Oxf) 2001, 54: 791-6.

8. Paisley AN, Banerjee M, Rezai M, et al. Changes in arterial stiffnessbut not carotid intimal thickness in acromegaly. J Clin EndocrinolMetab 2011, 96: 1486-92.

9. Hill JM, Zalos G, Halcox JP, et al. Circulating endothelial progen-itor cells, vascular function, and cardiovascular risk. N Engl J Med2003, 348: 593-600.

10. Vasa M, Fichtlscherer S, Aicher A, et al. Number and migratoryactivity of circulating endothelial progenitor cells inversely corre-late with risk factors for coronary artery disease. Circ Res 2001,89: E1-7.

11. Werner N, Kosiol S, Schiegl T, et al. Circulating endothelial pro-genitor cells and cardiovascular outcomes. N Engl J Med 2005,353: 999-1007.

12. Devin JK, Young PP. The effects of growth hormone and insulin-likegrowth factor-1 on the aging cardiovascular system and its pro-genitor cells. Curr Opin Invest Drugs 2008, 9: 983-92.

13. Devin JK, Vaughan DE, Blevins LS jr, et al. Low-dose growth hor-mone administration mobilizes endothelial progenitor cells inhealthy adults. Growth Horm IGF Res 2008, 18: 253-63.

14. Thum T, Hoeber S, Froese S, et al. Age-dependent impairment ofendothelial progenitor cells is corrected by growth-hormone-me-diated increase of insulin-like growth-factor-1. Circ Res 2007, 100:434-43.

15. Giustina A, Chanson P, Bronstein MD, et al. A consensus on cri-teria for cure of acromegaly. J Clin Endocrinol Metab 2010, 95:3141-8.

16. Esposito K, Ciotola M, Maiorino MI, et al. Circulating CD34+ KDR+endothelial progenitor cells correlate with erectile function and en-dothelial function in overweight men. J Sex Med 2009, 6: 107-14.

17. Esposito K, Maiorino MI, Di Palo C, et al. Effects of pioglitazoneversus metformin on circulating endothelial microparticles andprogenitor cells in patients with newly diagnosed type 2 diabetes- a randomized controlled trial. Diabetes Obes Metab 2011, 13:439-45.

Page 6: Circulating endothelial progenitor cells in acromegaly

G. Bellastella, M.I. Maiorino, R. Pivonello, et al.

830

18. Peichev M, Naiyer AJ, Pereira D, et al. Expression of VEGFR-2 andAC133 by circulating human CD34 cells identifies a population offunctional endothelial precursors. Blood 2000, 95: 952-8.

19. Salven P, Mustjoki S, Alitalo R, et al. VEGFR-3 and CD133 identifya population of CD34 lymphatic/vascular endothelial precursorcells. Blood 2003, 101: 168-72.

20. Cudmore MJ, Hewett PW, Ahmad S, et al. The role of het-erodimerization between VEGFR-1 and VEGFR-2 in the regulationof endothelial cell homeostasis. Nat Commun 2012, 3: 972.

21. Piecewicz SM, Pandey A, Roy B, Xiang SH, Zetter BR, Sengupta S.Insulin-like growth factors promote vasculogenesis in embryonicstem cells. PloS one 2012, 7: e32191.

22. Empen K, Lorbeer R, Völzke H, et al. Association of serum IGF1with endothelial function: results from the population-based studyof health in Pomerania. Eur J Endocrinol 2010, 163: 617-23.

23. Aicher A, Heeschen C, Mildner-Rihm C, et al. Essential role of en-dothelial nitric oxide synthase for mobilization of stem and pro-genitor cells. Nat Med 2003, 9: 1370-6.

24. Thum T, Fraccarollo D, Galuppo P, et al. Bone marrow molecularalterations after myocardial infarction: impact on endothelial pro-genitor cells. Cardiovasc Res 2006, 70: 50-60.

25. Asahara T, Murohara T, Sullivan A, et al. Isolation of putative pro-genitor endothelial cells for angiogenesis. Science 1997, 275: 964-7.

26. Urbich C, Dimmeler S. Endothelial progenitor cells: Characterizationand role in vascular biology. Circul Res 2004, 95: 343-53.

27. Schmidt-Lucke C, Rössig L, Fichtlscherer S, et al. Reduced numberof circulating endothelial progenitor cells predicts future cardio-vascular events: Proof of concept for the clinical importance of en-dogenous vascular repair. Circulation 2005, 111: b2981-7.

28. Fadini GP, Coracina A, Baesso I, et al. Peripheral blood CD34+KDR+endothelial progenitor cells are determinants of subclinicalatherosclerosis in a middle-aged general population. Stroke 2006,37: 2277-82.

29. Martin RM, Gunnell D, Whitley E, et al. Associations of insulin-likegrowth factor (IGF)-I, IGFII, IGF binding protein (IGFBP)-2 and IGF-BP-3 with ultrasound measures of atherosclerosis and plaque sta-bility in an older adult population. J Clin Endocrinol Metab 2008,93: 1331-8.

30. Kaplan RC, McGinn AP, Pollak MN, et al. Association of total in-sulin-like growth factor-I, insulin-like growth factor binding protein-1 (IGFBP-1), and IGFBP-3 levels with incident coronary events andischemic stroke. J Clin Endocrinol Metab 2007, 92: 1319-25.

31. Laughlin GA, Barrett-Connor E, Criqui MH, Kritz-Silverstein D. Theprospective association of serum insulin-like growth factor I (IGF-I)and IGF-binding protein-1 levels with all cause and cardiovasculardisease mortality in older adults: the Rancho Bernardo Study. JClin Endocrinol Metab 2004, 89: 114-20.

32. Akutsu H, Kreutzer J, Wasmeier G, et al. Acromegaly per se doesnot increase the risk for coronary artery disease. Eur J Endocrinol2010, 162: 879-86.

33. Menegazzo L, Albiero M, Avogaro A, Fadini GP. Endothelial pro-genitor cells in diabetes mellitus. BioFactors 2012, 38: 194-202.

34. Fadini GP, de Kreutzenberg SV, Coracina A, et al. CirculatingCD34+ cells, metabolic syndrome, and cardiovascular risk. EurHeart J 2006, 27: 2247-55.

35. Jialal I, Fadini GP, Pollock K, Devaraj S. Circulating levels of en-dothelial progenitor cell mobilizing factors in the metabolic syn-drome. Am J Cardiol 2010, 106: 1606-8.