Search

Chronic kidney diseases are accompanied by the accumulation of substances like asymmetric dimethylarginine, phenylacetic acid, homocysteine, and advanced glycation end products, known to either inhibit endothelial nitric oxide synthase (eNOS) or uncouple it, consequently limiting the amount of available nitric oxide (NO). Reduced bioavailability of NO induces endothelial dysfunction. An early loss of peritubular capillaries in tubulointerstitial fibrotic areas and injury to endothelial cells have been linked to progressive renal disease. Screening endothelial genes in cells treated with NOS inhibitors showed upregulation of collagen XVIII, a precursor of a potent antiangiogenic substance, endostatin. This finding was confirmed at the level of mRNA and protein expression. Tie-2 promoter-driven green fluorescent protein mice treated with nonhypertensinogenic doses of a NOS inhibitor exhibited upregulation of collagen XVIII/endostatin and rarefaction of capillary profiles. This was accompanied by the increased expression of transforming growth factor-β and connective tissue growth factor in the kidney. Occasional endothelial cells expressed both the marker of endothelial lineage (green fluorescent protein) and mesenchymal marker (α-smooth muscle actin or calponin). In vitro studies of endothelial cells treated with asymmetric dimethylarginine showed decreased expression of eNOS and Flk-1 and enhanced expression of calponin and fibronectin, additional markers of smooth muscle and mesenchymal cells. These cells overexpressed transforming growth factor-β and connective tissue growth factor, as well as endostatin. In conclusion, data presented here 1) ascribe to NO deficiency in endothelial cells the function of a profibrotic stimulus associated with the expression of an antiangiogenic fragment of collagen XVIII (endostatin) and 2) provide evidence of endothelial-mesenchymal transdifferentiation in the course of inhibition of NOS by a pathophysiologically important antagonist, asymmetric dimethylarginine. Both mechanisms may account for microvascular rarefaction.

We have previously found that nonenzymatically glycated collagen I (GC), mimicking diabetic microenvironment, can induce senescent phenotype in early passage human umbilical vein endothelial cells (HUVECs). In the present study, we explored the functional involvement of cell cycle checkpoint pathways in initiating GC-induced premature endothelial cell senescence. When compared with native collagen, early passage HUVECs showed increased p53, p21CIP1(p21), and p16INK4a(p16) mRNA expression after exposure to GC. Twenty-four hours after transfection of p16, p21, and p53-enhanced green fluorescent protein (EGFP) recombinant plasmids, HUVECs entered G1-phase cell cycle arrest. By days 3 and 5, HUVECs transfected with p16-EGFP showed an increased proportion of senescent cells, and this increase was more prominent in the GFP-positive cell population, which exhibited 68% of senescent cells. Transfection of p21 also induced senescence but only by day 5. Cotransfection of p16 and p21 showed no additive effect. Transfection of p21 or p53 induced apoptosis in HUVECs. Next, we suppressed endogenous p53, p21, p16, or retinoblastoma (Rb) gene expression through small interference RNA strategy and investigated their influence in p16- and p21-initiated endothelial cell senescence. Analysis indicated that suppression of p53 expression can abolish senescence induced by p16 overexpression. Paradoxically, this effect was not observed when p21 was suppressed. On the other hand, suppression of Rb eliminated senescence initiated by either p16 or p21 overexpression. In summary, the p53/p21 pathway is mainly responsible for GC-induced apoptosis, but the coordinated activation of the p53/p21 and p16 pathway is responsible for GC-induced endothelial cell senescence through a Rb-dependent mechanism.