1. Endothelial cell activation is correlated with increased cytosolic Ca¥ concentration, often monitored with cytoplasmic Ca¥ dyes, such as fura_2 and Calcium Green-1. We tested the hypothesis that during weak stimulation of porcine coronary artery endothelial cells, focal, subplasmalemmal Ca¥ elevations occur which are controlled by cell membrane Na¤-Ca¥ exchange near mitochondrial membrane and superficial endoplasmic reticulum (SER). 2. Bulk Ca¥ concentration ([Ca¥]b) was monitored using fura_2 or Calcium Green-1 and subplasmalemmal Ca¥ concentration ([Ca¥]sp) was determined with FFP-18. The distribution of the SER network was estimated using laser scanning and deconvolution microscopy. 1 nmol l¢ Bk or 10 mmol l¢ NaF yielded focal [Ca¥] elevation in the subplasmalemmal region with no increase in the perinuclear area. 6. Treatment with 10 ìmol l¢ nocodazole caused the SER to collapse and unmasked Ca¥ release in response to 1 nmol l¢ Bk and 10 mmol l¢ NaF, similar to low-Na¤ conditions, while the effect of thapsigargin was not changed. 7. These data show that in endothelial cells, focal, subplasmalemmal Ca¥ elevations in response to small or slow IP× formation occur due to vectorial Ca¥ release from the SER towards the plasmalemma followed by Ca¥ extrusion by Na¤-Ca¥ exchange. While these local Ca¥ elevations are not detectable with Ca¥ dyes for the determination of [Ca¥]b, prevention of Ca¥ extrusion or SER disruption yields increases in [Ca¥]b partially due to CICR. 8. All of the data support our hypothesis that in weakly stimulated endothelial cells, intracellular Ca¥ release and [Ca¥] elevation are limited to the subplasmalemmal region. We propose that the SER co-operates with associated parts of the plasma membrane to control Ca¥ homeostasis, Ca¥ distribution and Ca¥ entry. The existence of such a subplasmalemmal Ca¥ control unit (SCCU) needs to be considered in discussions of Ca¥ signalling, especially when cytoplasmic Ca¥ dyes, such as fura_2 or Calcium Green-1, are used.

1. Using single cell microfluorometry to monitor changes in bulk Ca¥ concentration ([Ca¥]bulk) and the whole-cell configuration of the patch clamp technique to measure K¤ currents (voltage clamp) and membrane potential (current clamp), the mechanisms of histamineinduced Ca¥ oscillations in the umbilical vein endothelial cell-derived cell line EA.hy926 were studied. 2. In single cells, histamine (10 ìÒ) evoked sinusoidal Ca¥ oscillations in low extracellular Ca¥ concentrations ([Ca¥]ï = 10-30 ìÒ). In contrast, histamine did not initiate Ca¥ oscillations either in the absence of extracellular Ca¥ (10 ìÒ EGTA) or in the presence of 2·5 mÒ extracellular Ca¥. 3. Ca¥ oscillations were accompanied by rhythmic activation of Ca¥-activated K¤ (KCa) channels and membrane hyperpolarization of 18·1 ± 3·9 mV. Hence, cell depolarization with 70 mÒ extracellular K¤ or the inhibition of non-selective cation channels (NSCCs) and KCa channels by 10 ìÒ Loe 908 and 10 mÒ tetrabutylammonium prevented histamine-evoked Ca¥ oscillations. 4. Preventing Na¤-Ca¥ exchange (NCX) by 10 ìÒ 2',4'-dichlorobenzamil, or removal of extracellular Na¤, abolished histamine-induced Ca¥ oscillations. Lowering the extracellular Na¤ concentration and thus promoting the reversed mode of NCX (3Na¤ out and 1Ca¥ in) increased the amplitude and frequency of histamine-induced Ca¥ oscillations by 25 and 13%, respectively. Hence, in the absence of extracellular Ca¥, 10 ìÒ histamine induced an elevation of intracellular Na¤ concentration in certain subplasmalemmal domains. 5. The inhibitor of sarcoÏendoplasmic reticulum Ca¥-ATPase (SERCA) 2,5-di-tert-butyl-1,4-benzo-hydroquinone (15 ìÒ) prevented histamine-induced Ca¥ oscillations. In addition, blockage of ryanodine-sensitive Ca¥ release (RsCR) by 25 ìÒ ryanodine blunted Ca¥ oscillations. 6. In endothelial cells that were treated for 16 h with 10 ìÒ nocodazole to collapse the superficial endoplasmic reticulum (sER), no histamine-induced Ca¥ oscillations were found. 7. We conclude that in low [Ca¥]ï conditions histamine-induced Ca¥ oscillations depend on transmembrane Na¤ loading through NSCCs that leads to Ca¥ entry via NCX. Cation influx is controlled by KCa channel activity that triggers membrane hyperpolarization and, thus, provides the driving force for cation influx. Hence, the Ca¥ entering needs to be sequestrated via SERCA into sER to become released by RsCR to evoke Ca¥ spiking. These data further support our previous work on localized Ca¥ signalling as a key phenomenon in endothelial Ca¥ homeostasis. Keywords:

The involvement of ryanodine‐sensitive Ca2+ release (RsCR) in bradykinin (Bk)‐induced Ca2+ release, capacitative Ca2+ entry (CCE) and nitric oxide synthase (NOS) activation was assessed in freshly isolated bovine coronary artery endothelial cells. Using deconvolution microscopy fura‐2 was found throughout the whole cytosol, while the cell membrane impermeable dye FFP‐18 was exclusively in the cell membrane. Thus, perinuclear ([Ca2+]pn) and subplasmalemmal Ca2+ concentration ([Ca2+]sp) were monitored using fura‐2 and FFP‐18. Inhibition of Na+−Ca2+ exchange by lowering extracellular Na+ concentration augmented the Bk‐induced [Ca2+]pn signal in Ca2+‐free solution. This effect was abolished when RsCR was prevented with 25 μmmu;mol l−1 ryanodine, while inhibition of RsCR had no effect on Bk‐induced increase in [Ca2+]pn without inhibition of Na+−Ca2+ exchange. Initiating RsCR by 200 nmol l−1 ryanodine increased [Ca2+]sp, while [Ca2+]pn remained constant. However, when Na+−Ca2+ exchange was prevented, ryanodine was also able to elevate [Ca2+]pn. Blockage of RsCR diminished Ca2+ extrusion in response to stimulation with Bk in normal Na+‐containing solution. Inhibition of RsCR blunted Bk‐activated CCE, while inhibition of Na+−Ca2+ exchange during stimulation enhanced CCE. Although direct activation of RsCR failed to activate NOS, inhibition of RsCR diminished the effect of ATP and Bk on NOS, while the effect of thapsigargin remained unchanged. These data suggest that during stimulation subplasmalemmal RsCR occurs, which contributes to the activities of CCE and NOS. Thus, the function of the subplasmalemmal Ca2+ control unit must be extended as a regulator for CCE and NOS.

In this study the contribution of alternating architecture and Ca2+ handling of mitochondria to cytosolic Ca2+ homeostasis was elucidated under normoglycemic and hyperglycemic (HGC) conditions in the human endothelial cell line EA.hy926. Exposure of endothelial cells to hyperglycemic medium elevated basal cytosolic free Ca2+ concentration ([Ca2+]cyto), the histamine-initiated cytosolic Ca2+ signaling, and the mitochondrial Ca2+ content after cell stimulation. The latter was possibly due to the prolonged mitochondrial Ca2+ elevation in response to agonists found in HGC-pretreated cells. Moreover, under HGC mitochondrial free radical production was increased and mitochondrial shape changed from a mainly tubular, highly interconnected network toward multiple, isolated singular structures. Such changes could not be correlated with HGC-induced alterations of cytosolic Ca2+ signaling that became normalized with antimycin A, an inhibitor of the respiratory chain. These data suggest that although mitochondrial structure changes considerably during HGC, alterations in cytosolic Ca2+ signaling are more likely due to the enhanced energy status/metabolism of the mitochondria. On the other hand, in normoglycemic cells of unforced fragmentation of mitochondria yielded elevated basal [Ca2+]cyto, while the global Ca2+ signaling in response to histamine remained unchanged. Thus, mitochondrial architecture (ie, tubular versus fragmented structure) per se does not have a detectable impact on agonist-initiated global cytosolic Ca2+ signaling, while this organelle represents an early target in hyperglycemia leading to alterations in cytosolic Ca2+ signaling.

Abstract-Although the involvement of free radicals in the development of endothelial dysfunction under pathological conditions, like diabetes and hypercholesterolemia, has been proposed frequently, there is limited knowledge as to how superoxide anions (O 2 Ϫ ) might affect endothelial signal transduction. In this study, we investigated the effects of preincubation with the O 2 Ϫ -generating system xanthine oxidase/hypoxanthine (XO/HX) on mechanisms for Ca 2ϩ signaling in cultured porcine aortic endothelial cells. Incubation of cells with XO/HX yielded increased intracellular Ca 2ϩ release and capacitative Ca 2ϩ entry in response to bradykinin and ATP in a time-and concentration-dependent manner. This effect was prevented by superoxide dismutase but not by the tyrosine kinase inhibitor tyrphostin A48. In addition, capacitative Ca 2ϩ entry induced by the receptor-independent stimulus 2,5-di-(tert-butyl)-1,4-benzohydroquinone or thapsigargin was enhanced in O 2 Ϫ -exposed cells (ϩ38% and ϩ32%, respectively). Increased Ca 2ϩ release in response to bradykinin in XO/HX-pretreated cells might be due to enhanced formation of inositol-1,4,5-trisphosphate (ϩ140%). Exposure to XO/HX also affected other signal transduction mechanisms involved in endothelial Ca 2ϩ signaling, such as microsomal cytochrome P450 epoxygenase and membrane hyperpolarization to Ca 2ϩ store depletion with thapsigargin (ϩ103% and ϩ48%, respectively) and tyrosine kinase activity (ϩ97%). A comparison of bradykinin-initiated intracellular Ca 2ϩ release and thapsigargin-induced hyperpolarization with membrane viscosity modulated by XO/HX (decrease in viscosity) or cholesterol (increase in viscosity) reflected a negative correlation between bradykinin-initiated Ca 2ϩ release and membrane viscosity. Because intracellular Ca 2ϩ is a main regulator of endothelial vascular function, our data suggest that O 2 Ϫ anions are involved in regulation of the vascular endothelium.