, Volume 50, Issue 10, pp 2156–2163

Glucose tolerance is negatively associated with circulating progenitor cell levels

  • G. P. Fadini
  • L. Pucci
  • R. Vanacore
  • I. Baesso
  • G. Penno
  • A. Balbarini
  • R. Di Stefano
  • R. Miccoli
  • S. de Kreutzenberg
  • A. Coracina
  • A. Tiengo
  • C. Agostini
  • S. Del Prato
  • A. Avogaro



Circulating progenitor cells participate in cardiovascular homeostasis. Depletion of the pool of endothelial progenitor cells (EPCs) is associated with increased cardiovascular risk. Furthermore, EPCs are reduced in the presence of classical risk factors for atherosclerotic disease, including diabetes mellitus. This study was designed to evaluate progenitor cell levels in volunteers with different degrees of glucose tolerance.


Cardiovascular parameters and the levels of circulating CD34+ and CD34+ kinase insert domain receptor (KDR)+ cells were determined in 219 middle-aged individuals with no pre-diagnosed alterations in carbohydrate metabolism. Glucose tolerance was determined by fasting and 2 h post-challenge glucose levels, with IFG and IGT considered as pre-diabetic states.


CD34+ and CD34+KDR+ cells were significantly reduced in individuals who were found to have diabetes mellitus, and were negatively correlated with both fasting and post-challenge glucose in the whole population. While only CD34+ cells, but not CD34+KDR+ cells, were significantly reduced in pre-diabetic individuals, post-challenge glucose was an independent determinant of the levels of both CD34+ and CD34+KDR+ cells.


Glucose tolerance was negatively associated with progenitor cell levels in middle-aged healthy individuals. Depletion of endothelial progenitors with increasing fasting and post-meal glucose may be one cause of the high incidence of cardiovascular damage in individuals with pre-diabetes.


Atherosclerosis Cardiovascular risk Diabetes Endothelium Glucose tolerance Pre-diabetes Progenitor cells Stem cells 



endothelial progenitor cell


kinase insert domain receptor


monoclonal antibody




Type 2 diabetes is associated with excess cardiovascular morbidity and mortality [1]. However, the true contribution of hyperglycaemia to cardiovascular risk remains controversial [2]. Agreement, on the other hand, exists on the cardiovascular risk preceding overt type 2 diabetes [3]. Thus in the Paris Prospective Study, incidence of CHD mortality was twice as high in IGT patients as in NGT individuals [4]. Similar results have been confirmed in European [5, 6], Asian [7] and Japanese [8] populations. However, the question of whether modest impairment in glucose homeostasis contributes independently to cardiovascular risk or instead represents a marker for a more complex condition carrying excess risk is still debated.

In recent years, much emphasis has been placed on elevation of postprandial glucose [9], although mechanisms linking postprandial hyperglycaemia to cardiovascular damage are still poorly understood [10]. Post-load hyperglycaemia is, for instance, more strongly associated with several cardiovascular risk factors than fasting hyperglycaemia [11]. These factors, together with the rapid surge in glucose concentrations, may cause endothelial injury and, over time, lead to atherogenic lesions. The development of atherosclerosis may be also dependent upon defective repair of the intimal layer, which relies on the bioavailability of circulating progenitor cells. Endothelial progenitor cells (EPCs) derive from bone marrow and can be mobilised to peripheral blood in response to many stimuli [12]. Once in the bloodstream, EPCs home to the sites of ischaemia and are physically integrated into the nascent vasculature, thus contributing to compensatory angiogenesis [13]. They also constitute a pool of cells that can form a patch at sites of endothelial damage, thus contributing to ongoing endothelial turnover and repair [14]. Classic cardiovascular risk factors are associated with reduction in circulating EPCs [15, 16]; currently, this is considered to be one mechanism by which risk factors influence vascular health [17], with EPCs now listed among surrogate indicators of risk [18]. We have demonstrated that type 2 diabetes is characterised by a profound EPC depletion and dysfunction [19, 20]. Moreover, in diabetic animals, inability to mobilise EPCs after ischaemia was associated with impaired compensatory angiogenesis [21], while replenishment of the EPC pool with mobilising agents or by cell transplantation restored vascular healing [22, 23]. Taken together, these data suggest that reduction and dysfunction of EPCs in diabetes mirror a poor endogenous regenerative capacity and favour the development of vascular complications.

On the other hand, little is known about potential EPC impairment in the early stages of the natural history of type 2 diabetes. We therefore tested the hypothesis that pre-diabetic states, such as IFG and IGT, are themselves associated with alterations in progenitor cell levels, which in turn may hamper the repairing activity in response to the endothelial damage that follows metabolic injury.


Study population

The study was approved by the local ethics committees and carried out in accordance with the Declaration of Helsinki as revised in 2000. Based on our previous data [15], we used the method of Dupont and Plummer [24] to calculate a sample size of around 200 individuals that would be able to detect a significant correlation between progenitor cell count and plasma glucose levels (type I error α = 0.05; power = 0.80; standard deviation of the regression error σ = 38; standard deviation of the independent variable σx = 71; predefined slope of the minimal linear correlation to be detected λ = 0.10). The same sample size was estimated to be sufficient to detect a statistically significant 12% difference in EPC count between control and pre-diabetic individuals (expected prevalence of pre-diabetes 0.25, with α = 0.05 and 1 − β = 0.80).

Individuals were recruited at the Metabolic Outpatient Clinics of the University of Padua (n =141) and University of Pisa (n = 78). Individuals participating in a screening programme for metabolic diseases and who had no sign of acute illness or infection, no immunological disease, no history of recent surgery, no uncontrolled hypertension and no established cardiovascular disease were recruited upon obtaining informed consent. All participants were at their first metabolic examination and none had any previously known abnormalities in glucose homeostasis. For all participants the following parameters were recorded: age, sex, family history for diabetes or cardiovascular disease, smoking habit (of one or more cigarettes per day), BMI, waist circumference and blood pressure (systolic and diastolic). Cardiovascular disease was ruled out using minimal criteria including personal history and physical examination.

After an overnight fast and abstention from smoking, an antecubital vein was cannulated for blood drawing. A basal blood sample was obtained for quantification of circulating EPCs and determination of lipid profile and fasting plasma glucose. After collecting basal blood samples, all participants underwent a 75 g OGTT for determination of plasma glucose 2 h after glucose load. Glucose tolerance was then defined according to the criteria set out by the American Diabetes Association [25].

Quantification of circulating endothelial progenitor cells

Peripheral blood progenitor cells were analysed for the expression of surface antigens with direct two- or three-colour flow cytometry (FACS Calibur; BD Biosciences, Franklin Lakes, NJ, USA) as previously reported [15, 19, 20]. To saturate sites for unspecific binding, before staining with specific monoclonal antibodies, cells were treated with fetal calf serum for 10 min and then washed with phosphate-buffered saline and 0.5% bovine albumin. Then, 150 μl peripheral blood were stained with 10 μl FITC-conjugated anti-human CD34 monoclonal antibody (mAb) (BD Biosciences) and 10 μl phycoerythrin (PE)-conjugated anti-human kinase insert domain receptor (KDR) mAb (R&D Systems, Minneapolis, MN, USA). Control isotype IgG1 and IgG2a antibodies were obtained from BD Biosciences. The frequency of peripheral blood cells positive for the above reagents was determined by a two-dimensional side scatter fluorescence dot plot analysis, after appropriate gating and staining with the different reagents: we initially gated CD34+ peripheral blood cells in the mononuclear cell fraction and then examined the resulting population for dual expression of KDR (Fig. 1). For FACS analysis, 5 × 105 cells were acquired and scored using a FACS Calibur analyser (BD Biosciences); cell counts are then expressed as number of cells per 106 cytometric events. Data were processed using a software program (Macintosh CELLQuest; BD Biosciences). The instrument set-up was optimised daily by analysing the expression of peripheral blood lymphocytes labelled with FITC anti-CD4, PE anti-CD8, PECy5 anti-CD3, and allophycocyanin anti-CD45 four-colour combination. Operators, who performed all tests, were trained in flow cytometry, practised with rare event analysis and blinded to the clinical status of participants.
Fig. 1

Representative flow cytometry scatter plots used to determine CD34+ and CD34+KDR+ cell counts. After morphological gating in the mononuclear cell fraction (a), CD34+ cells were identified using an anti-CD34 mAb (c) with respect to the negative isotype control (b). CD34+KDR+ cells were then identified in the CD34 gate using an anti-KDR mAb in the CD34+ gate (d). To show the extent of similarity between the two cell counts, CD34+ cell count was plotted against CD34+KDR+ cells on the ln scale (e) as well as in the natural scale (f). FSC forward scatter, SSC side scatter

Substrate measurement

Plasma glucose was measured on an analyser (Glucose Analyzer 2; Beckman Instruments, Fullerton, CA, USA) employing a glucose oxidase reaction. Plasma lipids were determined according to standard enzymatic procedures. HDL-cholesterol concentration was determined after precipitation of apolipoprotein B-containing lipoproteins with phosphotungstic acid/MgCl2 (Roche, Milan, Italy). LDL-cholesterol was calculated by the Friedewald formula. All determinations were performed with an autoanalyser (Hitachi/Roche, Milan, Italy).

Statistical analysis

Data are expressed as means±SD. CD34+ and CD34+KDR+ cell counts are expressed as cells per 106 events. However, due to skewed distribution, they were transformed in their natural logarithms for all statistical analyses, while plots display cell counts in the original unit of measure (cells per 106 events) on the ln scale. Differences between two or more groups were evaluated using Student’s t test and ANOVA, respectively, with the least significant difference post hoc test; the χ2 test was used for dichotomous variables. Correlations between cell counts and all cardiometabolic parameters were assessed with univariate regression analyses; parameters significantly correlated with cell counts were then entered in a multiple linear regression analysis, to assess their independent effects. Statistical significance was accepted at p ≤ 0.05.


Characteristics of the study population

Clinical and metabolic features of the study population are summarised in Table 1. Upon OGTT, 134 of 219 individuals had NGT, 61 (28%) had pre-diabetes (i.e. IFG defined as 6.1 mmol/l [110 mg/dl] ≤ fasting glucose < 7.0 mmol/l [126 mg/dl] or IGT defined as 7.8 mmol/l [140 mg/dl] ≤ 2 h glucose < 11.1 mmol/l [200 mg/dl] or a combination of the 2) and 24 (11%) had diabetes mellitus (defined as fasting glucose ≥ 7.0 mmol/l [126 mg/dl] or 2 h glucose ≥ 11.1 mmol/l [200 mg/dl]). None of the participants were taking medication at the time of the study. As can be appreciated from Table 1, participants with pre-diabetes, not unexpectedly, had an unfavourable cardiometabolic profile when compared with individuals with NGT, while diabetic participants had the worst profile.
Table 1

Characteristics of the study sample


Controls (n = 134)

Pre-diabetes (n = 61)

Diabetes (n = 24)

ANOVA p value

Age (years)

43.8 ± 7.9

51.7 ± 7.6a

55.0 ± 8.0a


Male (%)





Family history of diabetes (%)





Family history of CVD (%)





Smoking habit (%)





BMI (kg/m2)

24.8 ± 4.8

27.5 ± 4.5a

27.5 ± 4.8a


Waist circumference (cm)

89.7 ± 13.6

98.0 ± 11.4a

101.7 ± 11.4a


Systolic blood pressure (mmHg)

121.8 ± 13.5

128.1 ± 13.0a

137.5 ± 15.6a,b


Diastolic blood pressure (mmHg)

81.6 ± 9.9

83.2 ± 12.5

85.6 ± 11.9


Total cholesterol (mmol/l)

5.13 ± 0.91

5.47 ± 0.91a

4.97 ± 0.98b


LDL-cholesterol (mmol/l)

3.31 ± 0.92

3.67 ± 0.88a

3.15 ± 0.98b


HDL-cholesterol (mmol/l)

1.44 ± 0.43

1.39 ± 0.40

1.21 ± 0.37a


Triacylglycerol (mmol/l)

1.0 ± 0.55

1.52 ± 0.75a

2.12 ± 1.58a,b


Fasting glucose (mmol/l)

4.81 ± 0.52

5.87 ± 0.78a

7,42 ± 1.5a,b


2 h post-challenge glucose (mmol/l)

5.71 ± 1.16

8.32 ± 1.78a

13.74 ± 2.36a,b


CD34+ cells

431.8 ± 173.7

350.0 ± 209.7

284.2 ± 187.5


ln [CD34+]c

5.97 ± 0.46

5.74 ± 0.59a

5.45 ± 0.67a


CD34+KDR+ cells

78.5 ± 47.0

73.0 ± 53.5

45.6 ± 28.5


ln [CD34+KDR+]c

4.17 ± 0.70

4.02 ± 0.77

3.55 ± 0.88a,b


Values are presented as the means±SD.

aStatistically different vs control after adjusting for α-inflation (least significant difference post hoc test)

bStatistically different vs pre-diabetes after adjusting for α-inflation (least significant difference post hoc test)

cCell counts expressed as ln [CD34+] ± coefficient of variation

Circulating progenitor cells in individuals with carbohydrate metabolism abnormalities

Generic circulating CD34+ progenitor cells decreased as a function of glucose tolerance, as they were significantly reduced in participants with pre-diabetes as compared with participants with NGT (350.0 ± 209.7 vs 431.8 ± 173.7 [means±SD]; p = 0.002). When pre-diabetic individuals were compared with NGT individuals, those with IGT displayed a significant CD34+ cell count reduction (339.5 ± 195.0; p = 0.006), while those with IFG showed a nearly significant reduction (380.5 ± 252.5; p = 0.058). CD34+ progenitor cells were even lower in diabetic individuals, a further decrease that was nearly significant in comparison with pre-diabetic individuals (284.2 ± 187.5; p < 0.001 vs NGT and p = 0.057 vs pre-diabetes) (Table 1, Fig. 2).
Fig. 2

CD34+ (triangles) and CD34+KDR+ (circles) cell levels in subjects with NGT, IGT, IFG and diabetes mellitus (DM). Bars indicate normal range of cell count on the ln scale. *p < 0.05 vs NGT

Circulating CD34+KDR+ EPCs also showed a progressive decline from NGT, IFG, IGT and diabetic individuals (p < 0.001 for trend). Nonetheless, CD34+KDR+ cells were not significantly reduced in pre-diabetic individuals vs NGT (73.0 ± 53.5 vs 78.5 ± 47.0; p = 0.226), whereas a significant reduction was apparent in diabetic individuals (45.6 ± 28.5; p < 0.001 vs NGT and p = 0.016 vs pre-diabetes) (Table 1, Fig. 2).

When the lower limit for IFG was set at 5.6 mmol/l (100 mg/dl) instead of 6.1 mmol/l (110 mg/dl), the number of participants classified as IFG increased (n = 59) and the CD34+ cell count further decreased, becoming significantly lower vs NGT (370.5 ± 233.5 vs 427.6 ± 170.4; p = 0.007), while the CD34+KDR+ cell count remained almost identical.

Natural ln-transformed counts of circulating CD34+ and CD34+KDR+ progenitors correlated with both fasting (CD34+: r = −0.219; p = 0.001; CD34+KDR+: r = −0.144; p = 0.04) and post-challenge glucose (CD34+: r = −0.282; p < 0.001; CD34+KDR+: r = −0.212; p = 0.003) (Fig. 3).
Fig. 3

Linear correlations of CD34+ and CD34+KDR+ cell counts (original cell count on the ln scale) with fasting plasma glucose (FPG) (a, c) and post-load glucose (PLG) (b, d). r = −0.219, p = 0.001, residual coefficient of variation (rCV) = 0.62 (a); r = −0.282, p < 0.001, rCV = 0.69 (b); r = −0.144, p = 0.04, rCV = 0.59 (c); r = −0.212, p = 0.03, rCV = 0.58 (d)

In univariate regression analyses performed separately for all cardiometabolic parameters (see Table 2), CD34+ cell count was significantly negatively correlated with age, HDL-cholesterol, fasting plasma glucose and 2 h post-challenge plasma glucose. When those parameters were entered as explanatory variables in a multiple linear regression analysis, only HDL and post-challenge glucose remained significantly associated with CD34+ cell count.
Table 2

Univariate and multiple regression analyses to identify cardiometabolic variables significantly correlated with CD34+ and CD34+KDR+ cell counts (ln-transformed)

Explanatory variable

CD34+ cells

CD34+KDR+ cells





Age (years)

−1.19 (−0.42 to −1.96)*

−0.2 (−1.17 to 0.78)

−0.9 (−2.05 to 0.28)


13.1 (−2.37 to 31.0)

−14.27 (−30.5 to 5.73)

Family history DM

−2.37 (−15.9 to 13.3)

−7.32 (−25.0 to 14.5)

Family history CVD

−5.26 (−18.4 to 9.96)

−6.01 (−23.94 to 16.2)

Smoking habit

−1.69 (−19.0 to 19.4)

1.0 (−22.9 to 32.4)

Waist circumference (cm)

−0.1 (−0.69 to 0.49)

−0.7 (−1.47 to 0.08)

BMI (kg/m2)

−0.2 (−1.75 to 1.38)

−0.9 (−3.0 to 1.26)

Systolic BP (mmHg)

−0.1 (−0.69 to 0.49)

−0.7 (−1.47 to 0.08)

Diastolic BP (mmHg)

−0.1 (0.69 to 0.49)

−0.4 (−1.37 to 0.58)

Total cholesterol (mmol/l)

−0.3 (−8.19 to 8.26)

−2.96 (−13.4 to 8.73)

LDL-cholesterol (mmol/l)

−0.2 (−8.09 to 8.36)

−2.96 (−13.6 to 8.94)

HDL-cholesterol (mmol/l)

−23.9 (−35.9 to −9.6)*

−29.3 (−40.6 to −15.7)*

−0.3 (−22.3 to 29.4)

TG (mmol/l)

−0.4 (−8.63 to 8.57)

−3.73 (−70.9 to 218.2)

Fasting glucose (mmol/l)

−9.97 (−15.4 to −4.14)*

−4.02 (−12.8 to 5.66)

−9.15 (−17.0 to −0.58)*

−4.88 (−17.1 to 9.11)

2 h glucose (mmol/l)

−5.35 (−7.73 to −2.91)*

−5.26 (−8.36 to −2.05)*

−5.64 (−9.08 to −2.05)*

−4.59 (−9.15 to −0.20)*

Data are expressed as the relative percentage increase in the dependent variable per unit of increase in the explanatory variable (95%CI)

DM: diabetes mellitus, TG: triacylglycerol

*p < 0.05

In univariate regression, the CD34+KDR+ cell count was significantly correlated with fasting and post-challenge glucose. Upon multiple regression, only post-challenge glucose remained significantly associated with CD34+KDR+ cell counts (Table 2).


We demonstrate that circulating CD34+ and CD34+KDR+ progenitor cell counts are inversely related to glucose tolerance independently of concomitant risk factors. In light of the comprehensive role played by circulating progenitor cells in vascular homeostasis, exhaustion of these cells may contribute to the increased cardiovascular risk that characterises pre-diabetic conditions.

The notion that a subgroup of progenitor cells from the bloodstream has the ability to repopulate the endothelium has been substantiated by a vast literature of in vitro and animal studies [26]. EPCs can be isolated and expanded in culture to assess their properties and functional capacities: unfortunately, those costly and time-consuming procedures cannot be applied on a large scale. Therefore, in population-based studies, progenitor cells are defined and enumerated by flow cytometry, on the basis of the expression of surface antigens, a method considered the gold standard for this purpose [27]. Even if their exact phenotype is uncertain, in the present study, EPCs were defined as CD34+KDR+ cells, first because this is becoming the most widely accepted EPC phenotype [20, 28], and second because CD34+KDR+ cell count predicted cardiovascular outcomes in two large prospective studies [29, 30]. While CD34 is an adhesion molecule expressed mainly on haematopoietic stem cells, KDR (also known as type 2 vascular endothelial growth factor receptor) is a typical endothelial marker. In support of the hypothesis that CD34+KDR+ cells correspond to EPCs, it should be noted that freshly isolated human CD34+KDR+ cells showed vascular healing properties in animal models of coronary and peripheral artery disease [31, 32]. However, the simple evaluation of progenitor cells by flow cytometry implies the conceptual abstraction that antigenic phenotypes correspond to functional phenotypes, which is not always true: thus it is likely that CD34+KDR+ cells are a mixed population that includes EPCs. We also determined total CD34+ cells, as we have recently shown that this phenotype strongly correlates with all cardiometabolic parameters and provides an index of cardiovascular risk [15].

Regardless of the actual functional/antigenic correlations, both circulating CD34+ and CD34+KDR+ cells have been found to be reduced in subjects with established atherosclerotic disease [16, 33]. Previous studies have also shown that CD34+KDR+ cell count is reduced early on in pre-atherosclerosis and represents an independent predictor of carotid intima-media thickness in the general population [28]. These findings indicate that depletion of this putative EPC phenotype precedes the development of overt atherosclerosis, supporting the hypothesis that such alterations are not a mere disease epiphenomenon, but play a role in the process of atherogenesis [34].

Hyperglycaemic diabetic patients also showed a marked depletion of CD34+ and CD34+KDR+ cell pools, as well as a profound functional impairment of cultured EPCs [19, 20, 35]. This was hailed as a novel explanation for the extensive vascular damage in diabetic patients, also because EPC-based therapies restored vascular healing in diabetic animals [22, 23]. Less well known is the behaviour of these cells in the presence of minor abnormalities in glucose tolerance, and this despite the fact that IGT is known to be associated with increased cardiovascular risk [36, 37]. Indeed, alterations of glucose tolerance form a continuum, from NGT to overt diabetes, which progressively worsens vascular health [38]. We therefore recruited middle-aged healthy subjects at their first metabolic evaluation in order to explore the relation between progenitor cells and glucose tolerance. This approach had the advantage of studying a naïve population sample and minimising confounders, such as concomitant diseases and medications. Although, as noted before, we cannot infer that CD34+ and CD34+KDR+ cells perfectly correspond to EPCs, by showing a reduction of these cells in parallel with increases in fasting and post-load glucose, we do provide evidence that carbohydrate metabolism may regulate these progenitor cell pools. Therefore, even if the results of this study may not be directly extended to the general population, we suggest that the excess cardiovascular risk associated with pre-diabetes is likely to be sustained not only by the enhanced endothelial injury effected by hyperglycaemia and concomitant metabolic risk factors, but also by impaired endothelial regeneration resulting from depletion of circulating progenitors.

Interestingly, available experimental findings suggest a link between glucose toxicity and progenitor cell alterations. For instance, mild in vitro hyperglycaemia (12 mmol/l) reduced proliferation and increased apoptosis of cultured EPCs [39, 40]; additionally, reduction of plasma glucose levels in poorly controlled diabetic individuals was followed by an increase of circulating CD34+ progenitor cells [41]. In addition, we have demonstrated that stable hyperglycaemia impairs mobilisation of progenitor cells from bone marrow to peripheral blood in diabetic rats [21].

It is possible that clustering cardiovascular risk factors in patients with IGT contribute to reduce progenitor cell levels. Nonetheless, our data suggest that glucose tolerance as such, rather than the accompanying metabolic alterations, determined cell reduction. In fact, in the multiple analysis, the 2-h plasma glucose remained significantly associated with cell counts, regardless of other cardiovascular parameters and risk factors. Epidemiological and mechanistic studies have repeatedly suggested that increased 2 h post-OGTT and postprandial glucose may also confer great cardiovascular risk in pre-diabetic patients, as well as in diabetic patients [10, 42, 43]. Acute hyperglycaemia impairs endothelial function probably through induction of oxidative stress [44]. Although this was an association study not designed to prove cause–effect relationships, our demonstration that progenitor cells declined in parallel with rises in post-load glucose may indicate that excessive glycaemic elevations, as may occur after ingestion of a meal, reduce endothelial progenitors, also possibly via oxidative stress. The subsequent depletion of the EPC pool would result in the inability to maintain adequate vascular homeostasis and to stimulate compensatory angiogenesis where needed, ultimately leading to an elevated risk of cardiovascular events.

In the present study we have found that, as compared with individuals with NGT, CD34+ cells were significantly lower in pre-diabetic participants and further reduced in those with diabetes, suggesting that the decline in CD34+ progenitors follows the natural history of the development of diabetes, as well as the concomitant surge in cardiovascular risk. CD34+KDR+ cells also declined progressively with worsening glucose tolerance, but they were not significantly reduced in pre-diabetic individuals, as they were in diabetic individuals. Despite the cross-sectional nature of this study and the acknowledged difficulty of the antigenic-into-functional translation, it is tempting to speculate that depletion of generic CD34+ cells precedes reduction of the specific CD34+KDR+ population in the development of type 2 diabetes. This would agree with our observation that CD34+ does indeed represent the progenitor cell phenotype most closely correlated to cardiovascular risk [15]. Probably, the CD34+ population is very heterogeneous; moreover, it has been shown to include, besides haematopoietic stem cells, a wide range of endothelial, smooth muscle, perivascular and myocardial progenitor cells [45]. All these cellular compartments may be altered by the metabolic abnormalities of the pre-diabetic state, resulting in impairment of the endogenous regenerative potential of the whole cardiovascular system [46, 47]. Finally, it should be noted that when dealing with rare events such as circulating progenitor cells, single antigen flow cytometry proved to be more accurate and precise than double-antigen analysis [15]: this may be one reason for the non-significant results with the CD34+KDR+ phenotype.

In conclusion, we have shown that CD34+ and CD34+KDR+ cell counts are inversely related to glucose tolerance and that reduction of these putative endothelial progenitors may occur early in the natural history of type 2 diabetes, since CD34+ cells were already reduced in participants with pre-diabetes. Progenitor cell exhaustion appears to be influenced by the degree of post-OGTT hyperglycaemia, although the causal link responsible for such an association remains to be elucidated. Should our data reflect an actual deficiency of functional EPCs, we propose that the excess of cardiovascular risk in individuals with IGT derives not only from the endothelial insult triggered by multiple metabolic disturbances, but also from an impaired regenerative capacity of the endothelium. This concept would indicate possible novel therapeutic and preventive strategies aimed at improving endogenous endothelial healing.


Duality of interest

The authors declare that there is no duality of interest associated with this manuscript.


  1. 1.
    Almdal T, Scharling H, Jensen J, Vestergaard H (2004) The independent effect of type 2 diabetes mellitus on ischemic heart disease, stroke, and death: a population-based study of 13,000 men and women with 20 years of follow-up. Arch Intern Med 164:1422–1426PubMedCrossRefGoogle Scholar
  2. 2.
    UKPDS Group (1998) Intensive blood-glucose control with sulphonylureas or insulin compared with conventional treatment and risk of complications in patients with type 2 diabetes (UKPDS 33). Lancet 352:837–853CrossRefGoogle Scholar
  3. 3.
    Caballero AE (2005) Metabolic and vascular abnormalities in subjects at risk for type 2 diabetes: the early start of a dangerous situation. Arch Med Res 36:241–249PubMedCrossRefGoogle Scholar
  4. 4.
    Eschwege E, Richard J, Thibult N et al (1985) Coronary heart disease mortality in relation with diabetes, blood glucose and plasma insulin levels. The Paris Prospective Study, ten years later. Horm Metab Res Suppl 15:41–46PubMedGoogle Scholar
  5. 5.
    DECODE Study Group and the European Diabetes Epidemiology Group (2001) Glucose tolerance and cardiovascular mortality: comparison of fasting and 2-hour diagnostic criteria. Arch Intern Med 161:397–405CrossRefGoogle Scholar
  6. 6.
    Hu F, Stampfer M, Haffner S, Solomon C, Willett W, Manson J (2002) Elevated risk of cardiovascular disease prior to clinical diagnosis of type 2 diabetes. Diabetes Care 25:1129–1134PubMedCrossRefGoogle Scholar
  7. 7.
    Lawes C, Parag V, Bennett D et al (2004) Blood glucose and risk of cardiovascular disease in the Asia Pacific region. Diabetes Care 27:2836–2842PubMedCrossRefGoogle Scholar
  8. 8.
    Tominaga M, Eguchi H, Manaka H, Igarashi K, Kato T, Sekikawa A (1999) Impaired glucose tolerance is a risk factor for cardiovascular disease, but not impaired fasting glucose. The Funagata Diabetes Study. Diabetes Care 22:920–924PubMedCrossRefGoogle Scholar
  9. 9.
    Chiasson J, Josse R, Gomis R et al (2003) Acarbose treatment and the risk of cardiovascular disease and hypertension in patients with impaired glucose tolerance: the STOP-NIDDM trial. JAMA 290:486–494PubMedCrossRefGoogle Scholar
  10. 10.
    Ceriello A (2005) Postprandial hyperglycemia and diabetes complications: is it time to treat? Diabetes 54:1–7PubMedCrossRefGoogle Scholar
  11. 11.
    Yudkin J (2001) Post-load hyperglycaemia—an inappropriate therapeutic target. Lancet 359:166–167CrossRefGoogle Scholar
  12. 12.
    Asahara T, Murohara T, Sullivan A et al (1997) Isolation of putative progenitor endothelial cells for angiogenesis. Science 275:964–967PubMedCrossRefGoogle Scholar
  13. 13.
    Takahashi T, Kalka C, Masuda H et al (1999) Ischemia- and cytokine-induced mobilization of bone marrow-derived endothelial progenitor cells for neovascularization. Nat Med 5:434–438PubMedCrossRefGoogle Scholar
  14. 14.
    Iwakura A, Luedemann C, Shastry S et al (2003) Estrogen-mediated, endothelial nitric oxide synthase-dependent mobilization of bone marrow-derived endothelial progenitor cells contributes to reendothelialization after arterial injury. Circulation 108:3115–3121PubMedCrossRefGoogle Scholar
  15. 15.
    Fadini GP, de Kreutzenberg SV, Coracina A et al (2006) Circulating CD34+ cells, metabolic syndrome, and cardiovascular risk. Eur Heart J 27:2247–2255PubMedCrossRefGoogle Scholar
  16. 16.
    Vasa M, Fichtlscherer S, Aicher A et al (2001) Number and migratory activity of circulating endothelial progenitor cells inversely correlate with risk factors for coronary artery disease. Circ Res 89:E1–E7PubMedGoogle Scholar
  17. 17.
    Shantsila E, Watson T, Lip GY (2007) Endothelial progenitor cells in cardiovascular disorders. J Am Coll Cardiol 49:741–752PubMedCrossRefGoogle Scholar
  18. 18.
    Rosenzweig A (2005) Circulating endothelial progenitors—cells as biomarkers. N Engl J Med 353:1055–1057PubMedCrossRefGoogle Scholar
  19. 19.
    Fadini GP, Miorin M, Facco M et al (2005) Circulating endothelial progenitor cells are reduced in peripheral vascular complications of type 2 diabetes mellitus. J Am Coll Cardiol 45:1449–1457PubMedCrossRefGoogle Scholar
  20. 20.
    Fadini GP, Sartore S, Albiero M et al (2006) Number and function of endothelial progenitor cells as a marker of severity for diabetic vasculopathy. Arterioscler Thromb Vasc Biol 26:2140–2146PubMedCrossRefGoogle Scholar
  21. 21.
    Fadini GP, Sartore S, Schiavon M et al (2006) Diabetes impairs progenitor cell mobilisation after hindlimb ischaemia–reperfusion injury in rats. Diabetologia 49:3075–3084PubMedCrossRefGoogle Scholar
  22. 22.
    Jiao C, Fricker S, Schatteman G (2006) The chemokine (C–X–C motif) receptor 4 inhibitor AMD3100 accelerates blood flow restoration in diabetic mice. Diabetologia 49:2786–2789PubMedCrossRefGoogle Scholar
  23. 23.
    Schatteman GC, Hanlon HD, Jiao C, Dodds SG, Christy BA (2000) Blood-derived angioblasts accelerate blood-flow restoration in diabetic mice. J Clin Invest 106:571–578PubMedCrossRefGoogle Scholar
  24. 24.
    Dupont W, Plummer W (1990) Power and sample size calculations: a review and computer program. Control Clin Trials 11:116–128PubMedCrossRefGoogle Scholar
  25. 25.
    Akhter J (1997) The American Diabetes Association’s clinical practice recommendations and the developing world. Diabetes Care 20:1044–1045PubMedGoogle Scholar
  26. 26.
    Fadini GP, Sartore S, Agostini C, Avogaro A (2007) Significance of endothelial progenitor cells in subjects with diabetes mellitus. Diabetes Care 30:1305–1313PubMedCrossRefGoogle Scholar
  27. 27.
    Rustemeyer P, Wittkowski W, Jurk K, Koller A (2006) Optimized flow cytometric analysis of endothelial progenitor cells in peripheral blood. J Immunoassay Immunochem 27:77–88PubMedCrossRefGoogle Scholar
  28. 28.
    Fadini GP, Coracina A, Baesso I et al (2006) Peripheral blood CD34+KDR+ endothelial progenitor cells are determinants of subclinical atherosclerosis in a middle-aged general population. Stroke 37:2277–2282PubMedCrossRefGoogle Scholar
  29. 29.
    Werner N, Kosiol S, Schiegl T et al (2005) Circulating endothelial progenitor cells and cardiovascular outcomes. N Engl J Med 353:999–1007PubMedCrossRefGoogle Scholar
  30. 30.
    Schmidt-Lucke C, Rossig L, Fichtlscherer S et al (2005) Reduced number of circulating endothelial progenitor cells predicts future cardiovascular events: proof of concept for the clinical importance of endogenous vascular repair. Circulation 111:2981–2987PubMedCrossRefGoogle Scholar
  31. 31.
    Botta R, Gao E, Stassi G et al (2004) Heart infarct in NOD-SCID mice: therapeutic vasculogenesis by transplantation of human CD34+ cells and low dose CD34+KDR+ cells. FASEB J 18:1392–1394PubMedGoogle Scholar
  32. 32.
    Madeddu P, Emanueli C, Pelosi E et al (2004) Transplantation of low dose CD34+KDR+ cells promotes vascular and muscular regeneration in ischemic limbs. FASEB J 18:1737–1739PubMedGoogle Scholar
  33. 33.
    Kunz GA, Liang G, Cuculoski F et al (2006) Circulating endothelial progenitor cells predict coronary artery disease severity. Am Heart J 152:190–195PubMedCrossRefGoogle Scholar
  34. 34.
    Fadini GP, Sartore S, Agostini C, Avogaro A (2007) Endothelial progenitor cells and the natural history of atherosclerosis. DOI: 10.1016/j.atherosclerosis.2007.03.046 (in press)
  35. 35.
    Tepper OM, Galiano RD, Capla JM et al (2002) Human endothelial progenitor cells from type II diabetics exhibit impaired proliferation, adhesion, and incorporation into vascular structures. Circulation 106:2781–2786PubMedCrossRefGoogle Scholar
  36. 36.
    Meigs J, Nathan D, D’Agostino R Sr, Wilson P, Study FO (2002) Fasting and postchallenge glycemia and cardiovascular disease risk: the Framingham Offspring Study. Diabetes Care 25:1845–1850PubMedCrossRefGoogle Scholar
  37. 37.
    Bonora E (2002) Postprandial peaks as a risk factor for cardiovascular disease: epidemiological perspectives. Int J Clin Pract Suppl 129:5–11PubMedGoogle Scholar
  38. 38.
    Muntner P, Wildman R, Reynolds K, Desalvo K, Chen J, Fonseca V (2005) Relationship between HbA1c level and peripheral arterial disease. Diabetes Care 28:1981–1987PubMedCrossRefGoogle Scholar
  39. 39.
    Marchetti V, Menghini R, Rizza S et al (2006) Benfotiamine counteracts glucose toxicity effects on endothelial progenitor cell differentiation via Akt/FoxO signaling. Diabetes 55:2231–2237PubMedCrossRefGoogle Scholar
  40. 40.
    Krankel N, Adams V, Linke A et al (2005) Hyperglycemia reduces survival and impairs function of circulating blood-derived progenitor cells. Arterioscler Thromb Vasc Biol 25:698–703PubMedCrossRefGoogle Scholar
  41. 41.
    Humpert PM, Neuwirth R, Battista MJ et al (2005) SDF-1 genotype influences insulin-dependent mobilization of adult progenitor cells in type 2 diabetes. Diabetes Care 28:934–936PubMedCrossRefGoogle Scholar
  42. 42.
    Ceriello A (2004) Impaired glucose tolerance and cardiovascular disease: the possible role of post-prandial hyperglycemia. Am Heart J 147:803–807PubMedCrossRefGoogle Scholar
  43. 43.
    Cavalot F, Petrelli A, Traversa M et al (2006) Postprandial blood glucose is a stronger predictor of cardiovascular events than fasting blood glucose in type 2 diabetes mellitus, particularly in women: lessons from the San Luigi Gonzaga Diabetes Study. J Clin Endocrinol Metab 91:813–819PubMedCrossRefGoogle Scholar
  44. 44.
    Ceriello A, Taboga C, Tonutti L et al (2002) Evidence for an independent and cumulative effect of postprandial hypertriglyceridemia and hyperglycemia on endothelial dysfunction and oxidative stress generation: effects of short- and long-term simvastatin treatment. Circulation 106:1211–1218PubMedCrossRefGoogle Scholar
  45. 45.
    Yeh E, Zhang S, Wu H, Korbling M, Willerson J, Estrov Z (2003) Transdifferentiation of human peripheral blood CD34+-enriched cell population into cardiomyocytes, endothelial cells, and smooth muscle cells in vivo. Circulation 108:2070–2073PubMedCrossRefGoogle Scholar
  46. 46.
    Madeddu P (2005) Therapeutic angiogenesis and vasculogenesis for tissue regeneration. Exp Physiol 90:315–326PubMedCrossRefGoogle Scholar
  47. 47.
    Hirschi KK, Goodell MA (2002) Hematopoietic, vascular and cardiac fates of bone marrow-derived stem cells. Gene Ther 9:648–652PubMedCrossRefGoogle Scholar

Copyright information

© Springer-Verlag 2007

Authors and Affiliations

  • G. P. Fadini
    • 1
  • L. Pucci
    • 2
  • R. Vanacore
    • 2
  • I. Baesso
    • 1
  • G. Penno
    • 2
  • A. Balbarini
    • 2
  • R. Di Stefano
    • 2
  • R. Miccoli
    • 2
  • S. de Kreutzenberg
    • 1
  • A. Coracina
    • 1
  • A. Tiengo
    • 1
  • C. Agostini
    • 1
  • S. Del Prato
    • 2
  • A. Avogaro
    • 1
  1. 1.Department of Clinical and Experimental Medicine, Medical SchoolUniversity of PaduaPaduaItaly
  2. 2.Department of Endocrinology and Metabolism, Medical SchoolUniversity of Pisa, Hospital CisanelloPisaItaly

Personalised recommendations