Reduction of both number and proliferative activity of human endothelial progenitor cells in obesity

OBJECTIVE

Circulating endothelial progenitor cells (EPCs), responsible for neoangiogenesis and vascular repair, negatively correlate with vascular dysfunction and atherosclerotic risk factors. Because obesity may have a crucial role in the development of endothelial dysfunction, this study evaluated the number and proliferative activity of circulating human EPCs in obese (body mass index (BMI)=48+/-9, n=45) compared with lean (23+/-2, n=45) volunteers.

METHODS

EPCs were quantified after isolation of peripheral blood mononuclear cells (PBMCs) using fluorescence-activated cell sorting analyses. In addition, plated PBMCs developed colony-forming units (CFUs) from which 'outgrowth' endothelial cells (OECs) sprouted and differentiated into mature endothelial cells. Growth rates were monitored by periodical microscopic evaluation. Cell-cycle protein expression was determined by western blot analyses.

RESULTS

BMI negatively correlated (P<0.01) with the number of CD34(+)/CD133(+)/KDR(+) (r=-0.442), CD34(+)/KDR(+) (r=-0.500) and CD133(+)/KDR(+) (r=-0.282) EPCs. Insulin, leptin, HbA(1c), high-sensitivity C-reactive protein and hypertension, as well as diminished high-density lipoprotein and apolipoprotein A1, were not only associated with obesity but also with significantly reduced EPC levels. Applying selective culture conditions, EPC-CFUs differentiated into OECs that proliferated more slowly when derived from obese compared with lean subjects (obese: 19.9+/-2.2% vs lean: 30.9+/-3.2% grown area per week, P<0.01). The reduced proliferation was reflected by decreased (P<0.05, n=24 for each group) expression of cell-cycle-promoting cyclins and E2F-1, by hypophosphorylation of retinoblastoma protein and by increased (P<0.05, n=24 for each group) expression of the cell-cycle inhibitor p21(WAF-1/Cip1).

CONCLUSIONS

Reduced numbers of EPCs along with their premature senescence, as shown in this study, could function as early contributors to the development and progression of vascular dysfunction in obesity.

Insulin does not regulate glucose transport and metabolism in human endothelium

BACKGROUND

Although endothelial cells express insulin receptors, it is controversially discussed whether the endothelium represents an insulin-responsive tissue. Since available data are primarily restricted to animal endothelial cells, this study tested (i) whether insulin affects glucose metabolism in human endothelium; (ii) whether insulin sensitivity is different in micro- versus macrovascular endothelial cells; and (iii) whether glucose concentration in the incubation medium affects the cells' response to insulin.

MATERIALS AND METHODS

Human umbilical vein endothelial cells (HUVECs), human adult saphenous vein endothelial cells (HAVECs), human aortic endothelial cells (HAEC), and human retinal endothelial cells (HRECs) as well as human smooth muscle cells were incubated with/without insulin (0.3 nmol L(-1) or 1 micromol L(-1)). Glucose transport, glycogen synthesis, glycogen content, lactate release, and expression of phospho-Akt, Akt, and endothelial nitric oxide synthase (eNOS) were determined.

RESULTS

In HUVECs and HRECs, insulin (1 micromol L(-1)) increased (P < 0.05) eNOS expression by ~70% and doubled Akt phosphorylation, but the latter was by far more pronounced in human smooth muscle cells (+1093 +/- 500%, P < 0.05). In human smooth muscle cells, insulin (1 micromol L(-1)) stimulated glycogen synthesis by 67 +/- 11% (P < 0.01). In human micro- (HRECs) and macrovascular endothelial cells (HUVECs, HAVECs and HAECs), insulin, however, failed to stimulate glucose transport, glycogen synthesis, glycogen content, or lactate release under various conditions, i.e. after glucose deprivation or in medium with normal (5.5 mmol L(-1)) or high glucose (30 mmol L(-1)).

CONCLUSIONS

Insulin stimulated glycogen synthesis and Akt phosphorylation in human smooth muscle cells. In human micro- and macrovascular endothelial cells, insulin, however, failed to affect glucose uptake and metabolism under all experimental conditions applied, whereas it increased Akt phosphorylation and eNOS expression.

Thiazolidinediones inhibit proliferation of microvascular and macrovascular cells by a PPARgamma-independent mechanism

AIMS/HYPOTHESIS

This study evaluated the hypothesis that peroxisome proliferator-activated receptor-gamma (PPARgamma) agonists, including thiazolidinediones (TZDs) and the rexinoid LG100268 (LG), directly affect human vascular cell function (proliferation, cell cycle, protein expression, lactate release) independently of (1) their PPARgamma-activating potential and (2) the cells' vascular origin.

METHODS

Human umbilical vein endothelial cells (HUVECs), human adult vein endothelial cells (HAVECs), human retinal endothelial cells (HRECs) and human retinal pericytes (HRPYCs) were incubated (48 h) with 2-50 micromol/l rosiglitazone (RSG), RWJ241947 (RWJ), pioglitazone (PIO), troglitazone (TRO), 15-deoxy-Delta(12,14)-prostaglandin J2 (PGJ2) and LG. Proliferation, cell cycle distribution, protein expression, peroxisome proliferator-activated receptor responsive element (PPRE) transcriptional activity and mitochondrial effects were determined by [3H]thymidine incorporation, FACS analyses, western blots, reporter assays and lactate release respectively.

RESULTS

In HUVECs, RSG, RWJ, PIO, TRO, PGJ2 and LG reduced (p<0.01) proliferation (due to a G0/G1 cell cycle arrest) by up to 23%, 36%, 38%, 86%, 99% and 93% respectively. The antiproliferative response was similar in HRPYCs and HAVECs, but was attenuated in HRECs. Whereas p21WAF-1/Cip1 and p27Kip were differently affected in HUVECs, all agents reduced (p<0.05) expression of cyclins (D3, A, E, B), cyclin-dependent kinase-2 and hyperphosphorylated retinoblastoma protein. The rank order of the antiproliferative effects of TZDs in HUVECs (RSG approximately PIO approximately RWJ<TRO) contrasted their PPRE transcriptional activities (TRO<PIO<RSG<RWJ), but correlated with cellular lactate release. Proliferation inhibition and lactate release were mimicked by rotenone (mitochondrial complex I inhibitor).

CONCLUSIONS/INTERPRETATION

In conclusion, this study suggests that the antiproliferative action of the TZDs in vascular cells is independent of their PPARgamma-activating and associated insulin-sensitising potential, but could relate to mitochondrial mechanisms.

Modulation by leptin of proliferation and apoptosis in vascular endothelial cells

AIM

Plasma leptin concentrations not only correlate with body fat mass, but also with the degree of hypertensive retinopathy. The present study was designed to further examine, whether leptin's proliferative, proangiogenic activity relates to a yet uncovered anti-apoptotic effect.

RESULTS

Leptin (10-50 nmol/l) concentration-dependently reduced apoptosis in HUVECs (human umbilical vein endothelial cells), HAVECs (human adult vein endothelial cells) and HMECs (human microvascular endothelial cells) by 20% (P < or = 0.05). These findings were supported by increased expression of the apoptosis inhibitor bcl-2 (+55%, P < or = 0.05) as well as by differential modulation of the respective cell cycle checkpoint genes/proteins p53 (-20%, P < or = 0.01), p21(WAF-1/Cip1) (-23%, P < or = 0.05) and the Retinoblastoma protein (+123%, P < or = 0.01).

CONCLUSION

bcl-2 dependent anti-apoptotic action might contribute to leptin's proangiogenic activity and thereby promote the development of vascular proliferative disease in obesity.