Cellular function is largely determined by its structure. Thus, cellular deformability, a mechanical property of the cellular structural organization, varies in a number of physiological processes (including cell differentiation, growth and adhesion) and in pathological states (including oxidative stress, viral infection and, parasitic infection) . Recent studies have shown that alterations in endothelial cell membrane stiffness are accompanied with changes in NO production [15–17]. For example, a small physiological increase in extracellular sodium directly increases the stiffness of vascular endothelium and decreases NO release . An acute increase in potassium, within a physiological range, swells and softens endothelial cells and increases the release of NO . Nebivolol, a β1-receptor blocker, decreases membrane stiffness of endothelial cells, which is dependent on the increased content of NO outcome and is abrogated by L-NAME .
The above studies, however, have only focused on how experimental conditions altered endothelial cell membrane stiffness while simultaneously affecting NO synthesis. Thus, attention has not been paid to the exact effects of NO release and eNOS expression on endothelial cell membrane stiffness. Therefore, to examine the effect of NO on cell membrane stiffness, L-NAME was used to inhibit endogenous NO generation. Expression of eNOS (Figure 3), together with NO generation (as assessed by nitrite) (Figure 2), was quantified to clarify the relationship between cell membrane stiffness and NO metabolism. The data showed that L-NAME increased endothelial cell membrane stiffness (Figure 6), inhibited NO release and suppressed eNOS protein expression in a concentration-dependent manner suggesting a direct relationship between production of NO and the mechanical properties of the cell.
Endothelial cells subjected to a diabetic environment, both in vivo and in vitro, exhibit a diminished capacity for NOS-induced generation of NO . Consistent with this we observed that both high glucose and L-NAME similarly impaired eNOS expression and reduced NO production. Exposure to high glucose has been reported to impair NO metabolism through production of oxidative stress [5, 27, 28]. Superoxide may interfere with the generation of NO by several mechanisms including a decrease in endothelial eNOS expression mediated by activator protein AP-1, a change in the electrophysiological state of endothelial cells and the availability of tetrahydrobiopterin, an essential cofactor of eNOS. Interestingly hyperglycemia contributes to a switch in eNOS expression in a time-dependent manner. It was found that eNOS protein expression was significantly up-regulated 12 h following exposure to high glucose concentrations (30 mM), reaching a peak at 48h (two fold increase over baseline levels) . Pricci et al. also demonstrated that under high glucose conditions (20 mM), eNOS expression and nitrite/nitrate levels increase the first day, returning to normal levels at day three and diminishing thereafter . Consequently, under high glucose conditions, a compensatory increase in eNOS expression appears at this early stage, thereafter declining gradually as long-term oxidative stress develops. A deleterious effect of intermittent high glucose is also mediated by free radical over-production [3, 4, 29].
In agreement with previous studies, incubation of endothelial cells in constant or intermittent high glucose for seven days decreased not only eNOS expression (Figure 3), but also NO concentration in the culture media (Figure 2). Interestingly, there was no significant difference in NO production between the cells exposed to constant hyperglycemia and those incubated in fluctuating high glucose conditions (Figure 2). However, intermittent high glucose reduced eNOS expression more significantly than a constant level of hyperglycemia (Figure 3). This apparent discrepancy may be related to the observation that endothelial cells exposed to intermittent high glucose are more seriously impaired than those exposed to continuous high glucose due to enhanced oxidative stress [3, 4, 29]. Other reports show that high glucose increases eNOS protein expression, but ultimately leads to decreased NO release .
Hyperglycemia plays an important role in the etiology of endothelial dysfunction [26, 31]. Current data suggest that the deformability of endothelial cells detected by AFM will be affected in the presence of endothelial dysfunction [15–18, 21–23, 32, 33]. Despite this, few studies have used AFM to explore the effects of constant and intermittent high glucose on endothelial cell membrane elasticity. In this study, analysis of 300 force-distance curves obtained using AFM (Figure 5), showed that increased average membrane stiffness of endothelial cells (in response to stable or intermittent high glucose) accompanied decreased NO release and eNOS expression (Figures 2 and 3). Simultaneously, no significant difference existed in membrane stiffness and NO release between continuous and intermittent high glucose groups (Figures 2 and 6). Thus, it is suggested that the increased membrane stiffness of endothelial cells may be affected by the change of NO release more than that of eNOS expression.
In addition to the deleterious effects on endothelial NO-dependent function, high glucose concomitantly increases extracellular osmolality, which may itself impact cellular membrane stiffness. Thus, maintained and intermittent hyperosmolality (as caused by mannitol) was used as an osmotic control for continuous and intermittent high glucose. In the presence of mannitol the release of NO (Figure 2) was unaffected while the expression of eNOS (Figure 3) was slightly decreased. Although abnormal levels of osmolality are likely harmful to endothelial cells, the absence of a NO releasing effect presumably explains why hypertonicity was not shown to increase cell membrane stiffness (Figure 5). In fact, consistent with previous observations, hypertonicity decreased membrane stiffness .
Actin is not only an essential component necessary for maintenance of cellular integrity and function (e.g., membrane polarity, tight junctions, cellular adhesions, and signal transduction), but also undergoes dynamic changes in response to physiological and pathological stresses, including shear stress, vascular pressure and harmful mediators . Several studies have shown how decreases in NO synthesis and eNOS expression induce the redistribution of actin, especially F-actin [9–11, 35, 36]. NO-induced changes in F-actin filaments are proposed to be associated with mobilization of intracellular Ca2+ mediated by the cGMP-dependent pathway, which is activated by cGMP-dependent protein kinase G (PKG) . The stronger F-actin polymerization is also observed when endothelial cells are exposed to D-glucose pre-treated aortic smooth muscle cells, while D-manntinol has no effect on endothelial Ca2+ signaling . TGF-β, a growth factor closely linked to diabetic microvascular complications can stimulate F-actin assembly via activation of NADPH oxidase, which is a mechanism implicated in hyperglycemia . Overwhelming evidence demonstrates that actin redistribution can also regulate NO synthesis and eNOS activity not only through pre-translational mechanisms , but also through posttranslational mechanisms .
Consistent with the aforementioned studies, changes in F-actin relative fluorescence intensity in response to different media (Figures 6 and 7) were associated with decreased NO production (Figure 2) and decreased expression of eNOS (Figure 3). Treatment of endothelial cells with differing concentrations of L-NAME, stable and intermittent high glucose enhanced the apparent thickness of the longitudinal F-actin filaments as demonstrated by an increase in the relative fluorescence intensity of F-actin. Exposure to maintained and intermittent hyperosmolality did not alter NO production nor was, the fluorescence intensity of F-actin significantly different from the control group.
F-actin, as one of the major cytoskeletal components, clusters to form actin filaments, which are bundled and crosslinked by several actin-binding proteins into a network. The actin network plays a major role in determining the mechanical properties of living cells . Furthermore, evidence suggests that depolymerization/polymerization of F-actin filaments results in a dramatic decline/enhancement in endothelial cell membrane stiffness. For example, studies by Cuerrier have demonstrated that laturculin A, an F-actin filament depolymerizing agent, causes a dramatic decline in endothelial cell membrane stiffness . The variation in membrane elasticity in the various regions of endothelial cells (perinuclear or cytoplasmic membrane) is related to the distribution of cytoskeletal elements within these regions . In this study, it was demonstrated that the increased average membrane stiffness of endothelial cells treated with stable/intermittently high glucose and L-NAME was consistent with the change in the corresponding F-actin cytoskeleton (Figures 5, 6 and 7). Therefore, it is speculated that constant and intermittent high glucose, as well as L-NAME treatment, may stiffen endothelial cell membranes by an alteration in F-actin expression and arrangement through the described dysfunction of NO synthesis.