Constant or fluctuating hyperglycemias increases cytomembrane stiffness of human umbilical vein endothelial cells in culture: roles of cytoskeletal rearrangement and nitric oxide synthesis

Abstract

Background: Previous studies have implicated continuous or intermittent hyperglycemia in altered endothelium-derived nitric oxide (NO) synthesis. NO can regulate both the F-actin cytoskeleton and endothelial cell membrane stiffness. Atomic force microscopy (AFM) is a powerful tool that can be used to study plasma membrane deformability at the single cell level. As membrane stiffness is partially dependent on filamentous F-actin, the interdependence of these parameters can be studied through the combined approaches of AFM and laser scanning confocal microscopy (LSCM). In the present study, we evaluated the effects of constant or fluctuating hyperglycemia on endothelial-derived NO synthesis, the cytoskeletal contribution and endothelial cell membrane stiffness.

Results: Compared to control cells cultured in low glucose (5 mM), constant (25 mM) or fluctuating (25/5 mM) high glucose significantly decreased NO release along with stiffening of endothelial cell membranes and F-actin rearrangement. The non-selective nitric oxide synthase (NOS) inhibitor, N(G)-nitro-(L)-arginine methyl ester ((L)-NAME) exerted similar effects on endothelial cells. Increasing concentrations of (L)-NAME (from 0.1 to 1 mM) exacerbated these effects in a concentration-dependent manner.

Conclusions: Result from the present study suggest that stiffening endothelial cell membranes are associated with decreased NO synthesis, which was established through the F-actin cytoskeletal redistribution. The precise mechanisms of hyperglycemia-induced endothelial dysfunction require further investigation.

Figure 1

Force-versus-distance curve. Point A of the curve indicates the probe connects with the surface of the cells. From point A to B, the probe presses on the cell membrane and the opposing force generated along with increasing strain in the cell surface rises. At point B, the probe is ready to detach from the surface of the cell and no longer pushes against the membrane, so the opposing force reaches peak.

Figure 2

Effects of high glucose and the NOS inhibitor _L-NAME on the release of NO in human umbilical vein endothelial cells. NO concentrations in the culture supernatant were measured seven days after incubation of endothelial cells in various media. Control, continuous low glucose (5mM) media; HG, continuous high-glucose (25mM) media; FG, fluctuating glucose (25/5mM, alternated every 24h) media; HM, continuous low glucose (5mM) media containing 20mM mannitol; FM, low glucose (5mM) media with 0/20mM mannitol (alternated every 24h); 0.1mM, 0.5mM and 1mM, continuous low glucose (5mM) media containing 0.1, 0.5 or 1mM _L-NAME, respectively. Data shown in the figure are mean ± SD. “*”, P<0.05, compared to the control group. “#”, P<0.05, compared to the constant high glucose group.

Figure 3

Effects of high glucose and the NOS inhibitor _L-NAME on the expression of eNOS in human umbilical vein endothelial cells. The detection of the protein expression was carried out seven days after incubation of endothelial cells in various media (as described in Figure 2). (A) Representative western immunoblots of eNOS and β-actin. (B) Relative expression levels of eNOS proteins obtained using densitometric analyses of the blots, compared with those of β-actin. Data shown in the figure are mean ± SD. “*”, P<0.05, compared to the control group. “#”, P<0.05, compared to the constant high glucose group.

Figure 4

The topological morphology (A-B) and the imaging of the third dimension (C) of endothelial cell membrane detected using AFM. B and C were zoomed in from the region a in A. The areas b and c in B show the peripheral and central regions in the cell, where the force-distance curves were detected through the indentation force along the alteration of the distance between the tip and the studied location. Scale bar in A = 20μm; Scale bar in B-C = 10μm. The histogram (D) representing the morphologic characteristic was obtained from the scan of the cellular surface (B and C) with AFM. The force-distance curve (E) was obtained with the tip on the cell treated with continuous low glucose for seven days and presented the relative elasticity of the cell surface (approach curve, 2→3), with a noticeable adhesion force (retraction curve, 3→4).

Figure 5

Effects of high glucose and the NOS inhibitor _L-NAME on the average membrane stiffness of the whole cell. Measurement of average membrane stiffness was performed after seven days of incubating endothelial cells in various media (as described in Figure 2). The average membrane stiffness of five cells in each group was assessed through the slope of the ascending part of 300 force-distance curves detected using AFM. Data shown in the figure are mean ± SD. “*”, P<0.05, compared to the control group. “#”, P<0.05, compared to the constant high glucose group.

Figure 6

Effects of high glucose and the NOS inhibitor _L-NAME on F-actin cytoskeleton in human umbilical vein endothelial cells. The detection of F-actin fluorescence was performed following seven days of incubating endothelial cells in various media (as described in Figure 2). F-actin was probed with Rhodamine Phalloidin to emit red fluorescence. Scale bar = 50μm.

Figure 7

Comparisons of the relative fluorescence intensity of F- actin in human umbilical vein endothelial cells incubated in various media (as described in Figure 2) for seven days. Data shown in the figure are mean ± SD. “N=50”, the average F-actin fluorescence intensity from 50 fluorescent-staining endothelial cells each group, “*”, P<0.05, compared to the control group. “#”, P<0.05, compared to the constant high glucose group.