INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

VASOACTIVE SUBSTANCES and hemodynamic forces greatly affect the diameter of arterioles and hence peripheral vascular resistance. The arteriolar endothelium has a pivotal role in mediation of certain agonist-induced dilations and in sensing and transmitting alterations of hemodynamic forces into vasomotor responses (20). We (21) have shown previously that arteriolar dilation to acetylcholine (ACh) and to increases in intraluminal flow-related wall shear stress (WSS) is mediated by synthesis of endothelial factors.

It is generally believed that synthesis/release of endothelial mediators is controlled and preceded by changes in endothelial cell intracellular calcium concentration ([Ca²⁺]_i). Several studies on cultured endothelial cells (5), isolated arteries (19, 27) and arterioles (8, 12, 31) demonstrated that after application of endothelium-dependent dilator agents, such as ACh, there is a substantial increase in endothelial [Ca²⁺]_i, which is believed to be necessary to activate endothelial nitric oxide (NO) synthase (eNOS), leading to NO release and dilation. Previous studies in isolated arterioles provided indirect evidence for this idea showing that the Ca²⁺ ionophore A-23187, which promotes receptor-independent entry of Ca²⁺ into endothelial cells, releases NO and/or prostaglandins to dilate arterioles (16).

Studies on cultured endothelial cells also revealed some of the possible pathways that may cause activation of eNOS without a significant increase in [Ca²⁺]_i (10, 13, 26, 28), and thus it was hypothesized that such pathways may also be involved in vivo in the signal transduction of shear stress. Although an increase in endothelial [Ca²⁺]_i to agonists is believed to be a potent activator of eNOS, the possible role of Ca²⁺ signaling in the endothelial signal transduction of shear-stress-induced activation of eNOS in the endothelium of intact microvessels is not yet established. The contradictory findings are most likely due to the different experimental approaches used. In cultured endothelial cells, an increase in shear stress was shown to elicit increases in [Ca²⁺]_i (14) and release of NO (23). However, in these studies, there is a possibility that the culture solution contained vasoactive agents such as ATP, which also affects [Ca²⁺]_i and thus could interfere with determination of shear-stress-induced changes in [Ca²⁺]_i in endothelial cells (7, 29). Also, it can be assumed that level of shear stress and expression of relevant proteins and cytoskeletal structures involved in mechanotransduction differ between endothelial cells in culture and that of intact pressurized vessels (23, 24).

As of today, few studies have investigated agonist- and flow-induced changes in endothelial [Ca²⁺]_i in intact arterioles. In cremaster arterioles, in which prostaglandins are the primary mediators (22), an increase in endothelial [Ca²⁺]_i to initiation of intraluminal flow that was reported (8). In contrast, in coronary arterioles, only minor changes in endothelial [Ca²⁺]_i were reported (31) that was in response to increases in shear stress (1-5 dyn/cm²) followed by NO-mediated dilation raising the possibility that in this microvascular bed simultaneously operating Ca²⁺-independent pathways are also involved in the flow-induced activation of NO synthesis. The existence of this pathway is supported by recent studies on cultured endothelial cells identifying a novel mechanism for Ca²⁺-independent phosphorylation and activation of eNOS by tyrosine kinases (9, 10, 13, 26, 28).

It is of note that in intact vessels, increases in shear stress during increases in flow velocity elicit rapid dilation (delay time ~10 s). The resultant increase in the diameter of the vessel instantly reduces shear stress, thereby significantly reducing the signal sensed by the endothelium (20), as well as the changes in signaling pathway(s) and, if any, the elevation of [Ca²⁺]_i. Moreover, the magnitude and characteristics of shear stress-induced changes in endothelial [Ca²⁺]_i may be significantly affected by changes in vascular smooth muscle tone and the level of [Ca²⁺]_i, a mechanism recently suggested (6). Thus to exclude the interference of these mechanisms and to study the sole effect of agonists and shear stress on [Ca²⁺]_i, we performed experiments in which the arteriolar diameter was "clamped" at a constant level by maximally dilating arterioles.

The aim of our study was to elucidate the role of changes in endothelial [Ca²⁺]_i and the possible contribution of Ca²⁺-independent pathways in NO-mediated dilation in response to agonists and increases in shear stress. We hypothesized that both Ca²⁺-dependent and -independent pathways are present in the endothelium of skeletal muscle arterioles but that they are activated differently by agonists and shear stress.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Isolation of arterioles, videomicroscopy. Experiments were conducted on isolated arterioles of rat gracilis muscle of 12-wk-old Wistar rats (n = 30) as previously described (37). Intraluminal flow was established at a constant intraluminal pressure (80 mmHg) by changing inflow and outflow pressures to an equal degree but in opposite directions to keep constant midpoint luminal pressure. Perfusate flow was measured with a ball flowmeter (Omega Engineering).

Measurement of endothelial [Ca²⁺]_i. The [Ca²⁺]_i in arteriolar endothelium was assessed by the ratiometric fura 2 fluorescence method (19, 31, 37) using the Ionoptix Microfluorimeter System of videomicroscopy (Ionoptix; Milton, MA). This optical system measures the emitted fluorescence intensities of the Ca²⁺-sensitive fluorescent dye fura 2, the ratio of which was shown to correlate with the concentration of free Ca²⁺. The background signal from the cannulated arteriole was measured, and then the arteriolar endothelium was loaded with fura 2 by intraluminal perfusion of a physiological salt solution containing 5 µM fura 2-AM (10 min at 37°C) and 0.01% pluronic F-127 using a pressurized loading loop (19). During this procedure, the arteriole was illuminated with ultraviolet light and the emitted fluorescent signal was continuously monitored. Entry of fura 2 dye in the lumen was indicated by a rapid rise in emitted fluorescence intensities at each excitation wavelength. After the loading period, the lumen was flushed with fura 2-free physiological salt solution for 10 min. Endothelial calcium fluorescence ratios (ER_Ca), estimates of [Ca²⁺]_i, were calculated as previously described (37). To determine whether selective loading of the endothelium was accomplished, at the end of the experiments the endothelium was removed by perfusion of the vessel with air (37). The efficacy of endothelial denudation was ascertained by the lack of arteriolar responses to ACh after the administration of the air bolus. Selective endothelial loading was considered successful, if the following criteria were met (19): 1) removal of the fura 2-loaded endothelium of arterioles eliminated 90% of the fluorescence signal, and 2) 80 mmol/l KCl elicited constriction without an increase in ER_Ca. Only data obtained in arterioles that met these criteria were used. The protocol used produced selective and consistent loading of the endothelium in ~75% of arterioles.

Experimental protocols. After the equilibration period, arteriolar dilation to increases in shear stress (from 0 to ~20 dyn/cm² in control, by increasing intraluminal flow), to the Ca²⁺ ionophore A-23187 (10⁷ to 10⁶ mol/l), acetylcholine (ACh, 10⁹ to 10⁶ mol/l), and histamine (10⁶ to 10⁴ mol/l) were measured. The arterioles were then incubated for 30 min with N-nitro-L-arginine methyl ester (L-NAME, 10⁴ mol/l), a selective inhibitor of NOS, and arteriolar responses were reassessed. Indomethacin (10⁵ mol/l), a cyclooxygenase inhibitor, was continuously present during some of the experiments.

6

6

8

7

6

6

3

Materials. Fura 2-AM (Molecular Probes) and pluronic F-127 (Biomol; Plymouth, PA) were used. All other salts and chemicals were obtained from Sigma (St. Louis, MO). Fura 2-AM was dissolved in DMSO as a 1 mmol/l stock solution and frozen in 5-µl aliquots and protected from light until used. Each drug was added to the organ chambers, and final concentrations are reported.

Data analysis. WSS was calculated from diameter (2r) and flow data according to the equation WSS = 4Q/r³, where  is the viscosity of perfusate (0.007 poise at 37°C), Q is the perfusate flow, and r is the vessel radius (21). Changes in ER_Ca are expressed as a percentage of the baseline (37). All data are expressed as means ± SE. Statistical analyses were performed by analysis of variance. P values of < 0.05 were considered statistically significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Arteriolar dilations to shear stress, ACh, and histamine. Isolated gracilis muscle arterioles developed spontaneous myogenic tone (~40% of passive diameter) in response to 80 mmHg of intraluminal pressure without the use of any vasoactive agent. The passive diameter of arterioles was 150 ± 14 µm. First, we confirmed that increases in flow from 0 to 50 µl/min corresponding to shear stress of 0 to ~20 dyn/cm² elicit gradual dilations, reaching the maximum flow-induced response in intact arterioles of this size (Fig. 1A). Increases in shear stress above 10 dyn/cm² did not elicit further dilations. In the presence of indomethacin flow-induced arteriolar dilation was completely inhibited by L-NAME (Fig. 1A). ACh and histamine also elicited substantial arteriolar dilations, which were significantly inhibited by L-NAME (Fig. 1B).

View larger version (23K): [in this window] [in a new window]   Fig. 1.   A: arteriolar dilations to increases in shear stress in the absence and presence of the nitric oxide (NO) synthase (NOS) inhibitor N-nitro-L-arginine methyl ester (L-NAME). B: arteriolar dilations to acetylcholine (ACh) and histamine in the absence and presence of L-NAME. Indomethacin (Indo) was continuously present throughout the experiments (n = 8 arterioles). *P < 0.05. C: representative changes in the endothelial fluorescence ratio (ER340/380) recorded from the fura 2-loaded endothelium of an isolated intact skeletal muscle arteriole in response to initiation of intraluminal flow (corresponding to a shear stress value of 18 dyn/cm2) and the addition of ACh. In this experiment, ACh was applied during the flow experiment to show that endothelial intracellular calcium concentration ([Ca2+]i) can substantially increase if an appropriate stimulus is used. wo, Washout.

Changes in ER_Ca to shear stress, ACh, and histamine. Figure 1C shows representative changes in the fluorescence ratio recorded from the fura 2-loaded endothelium of an isolated intact skeletal muscle arteriole as a function of time in response to initiation of intraluminal flow and the addition of ACh to the organ bath. Verapamil was used to maximally dilate endothelium-loaded arterioles ("diameter clamped"), which did not affect ER_Ca responses. ACh and histamine in a concentration-dependent manner elicited substantial increases of ER_Ca (Fig. 2A). Increases in shear stress in the range of 0 to ~20 dyn/cm², which were shown to elicit maximal NO-mediated dilations (Fig. 1A), elicited minimal changes in ER_Ca (Fig. 2B). There was only a weak correlation between shear stress values and changes in ER_Ca (r² = 0.5) at lower values of (up to 20 dyn/cm²), and even higher values of shear stress (up to ~100 dyn/cm²) did not elicit further increases in endothelial [Ca²⁺]_i. These data are summarized in Fig. 2B. In addition, the time course of the responses elicited by ACh and shear stress differed greatly from each other. ACh elicited a biphasic response with a rapid peak (maximum at ~10 s) that was followed by a plateau phase, whereas responses to shear stress were monophasic and developed much slower (maximum at >60 s, Fig. 2C). In the absence of Ca²⁺ in the perfusate solution, there was no change in ER_Ca to a step increase in shear stress (Fig. 2D), whereas ACh still elicited a monophasic peak rise (maximum ~25%) that returned to baseline within ~40 s (Fig. 2D).

View larger version (24K): [in this window] [in a new window]   Fig. 2.   A and B: changes in endothelial [Ca2+]i (expressed as ERCa) to ACh, histamine, and wall shear stress in isolated arterioles. C and D: time course of changes in ERCa to an increase in shear stress (by initiation of intraluminal flow, corresponding to ~10 dyn/cm2) or to the administration of ACh (106 mol/l) in the presence and absence of Ca2+ in the perfusate solution (n = 6-20 arterioles).

Responses to A-23187. A-23187 elicited significant dilations that were abolished by L-NAME (Fig. 3A). A-23187 also elicited substantial increases in ER_Ca in diameter-clamped arterioles (Fig. 3B) with maximal changes (38 ± 5%) similar in magnitude to those obtained in response to ACh. However, responses to A-23187 were monophasic and reached the maximum in >6 min. In the absence of Ca²⁺ in the perfusate solution, increases in ER_Ca to A-23187 were absent (Fig. 3B). The magnitude of NO-mediated arteriolar dilations to shear stress, A-23187, histamine, and ACh were calculated by subtraction of responses obtained in the presence of L-NAME from the control values. NO-mediated dilations were then plotted against the changes in endothelial [Ca²⁺]_i to shear stress (5-10 dyn/cm²), A-23187 (10⁶ mol/l), histamine (10⁶ to 10⁴ mol/l), and ACh (10⁸ to 10⁶ mol/l) (Fig. 3C). We found a close correlation among the concentration of agents, increases in endothelial [Ca²⁺]_i, and the magnitude of dilations mediated by NO, indicating that these agonists elicit NO release predominantly by elevating endothelial [Ca²⁺]_i. In contrast, the relationship between shear-stress-induced changes in endothelial [Ca²⁺]_i and L-NAME-sensitive dilations yielded a steeper slope, indicating that increases in WSS elicit substantial NO-mediated dilations without substantial increases in endothelial [Ca²⁺]_i.

View larger version (17K): [in this window] [in a new window]   Fig. 3.   A: arteriolar dilations to the Ca2+ ionophore A-23187 in the absence and presence of the NOS inhibitor L-NAME. Indo was present throughout the experiments. *P < 0.05. B: time course of changes in endothelial [Ca2+]i (ERCa) to A-23187 (106 mol/l) in the presence and absence of Ca2+ in the perfusate solution. (n = 3-6 arterioles) C: relationships between the magnitude of changes in ERCa and L-NAME-sensitive, NO-mediated, dilations to increases in shear stress, A-23187, histamine, and ACh.

Effects of genistein on arteriolar responses to shear stress, ACh, and SNP. The flow-diameter relationship was also assessed in control conditions and after incubating the vessels for 30 min with intraluminal genistein, a specific tyrosine kinase inhibitor, or daidzein, an inactive form of genistein. In the presence of genistein but not daidzen, flow-induced dilation was completely eliminated (Fig. 4A); however, arteriolar dilations to ACh and SNP were not significantly affected (Fig. 4B).

View larger version (18K): [in this window] [in a new window]   Fig. 4.   Arteriolar dilations to increases in intraluminal flow (*P < 0.05) (A), 107 mol/l ACh or 107 sodium nitroprusside (SNP) (B) before and after the intraluminal administration of the tyrosine kinase inhibitor genistein (n = 3-6 arterioles). C: proposed scheme for endothelial Ca2+-dependent and -independent pathways for mediation of dilations to shear stress, A-23187, and ACh by NO in skeletal muscle arterioles. Presence of a non-NO, nonprostaglandin factor i.e., endothelium-derived hyperpolarizing factor (EDHF), is included on the basis of previous studies. eNOS, endothelial NOS.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The main findings of the present study are that in skeletal muscle arterioles NO-mediated dilations to increases in shear stress are associated with only slight increases in endothelial Ca²⁺, whereas to agonists (ACh, histamine, and A-23187) they are preceded by substantial increases in endothelial Ca²⁺. The Ca²⁺-independent dilation to shear stress can be blocked by the tyrosine kinase inhibitor genistein.

First, we confirmed our previous findings (21) that in skeletal muscle arterioles a significant portion of dilation to flow/shear stress is mediated by NO (Fig. 1A) and dilations to the Ca²⁺ ionophore A-23187 are mediated entirely by NO (16) (Fig. 3A), whereas dilations to ACh and histamine are mediated by a corelease of NO (Fig. 1B) and a non-NO, nonprostaglandin factor (2, 3, 17).

Next, we aimed to elucidate the role of endothelial [Ca²⁺]_i in the signal transduction of dilations to agonists and shear stress that are mediated by NO. The endothelium of diameter-clamped skeletal muscle arterioles was loaded intraluminally with the fluorescent dye fura 2, enabling us to detect changes in endothelial [Ca²⁺]_i to shear stress without interference by changes in vascular smooth muscle tone and [Ca²⁺]_i (6). Another important aspect of our study is that we investigated changes in endothelial [Ca²⁺]_i to step increases in shear stress without the interference of changes in diameter, because changes in diameter were prevented by maximal dilation of vessels.

We found that in intact skeletal muscle arterioles ACh, histamine (Fig. 2A), and A-23187 (Fig. 3B) elicited substantial increases in endothelial [Ca²⁺]_i. In contrast, increases in shear stress elicited only slight increases in endothelial [Ca²⁺]_i (Fig. 2B). Similar slight shear-stress-induced endothelial Ca²⁺ responses were reported in a recent study of intact coronary arterioles (31). Our study extended this finding to a higher range of shear stress corresponding to in vivo levels in arterioles of this size (25). It is also of note that the effective value of shear stress that elicits dilation varies significantly with the size of arterioles. In addition, our study also revealed important differences in the time kinetics of the endothelial Ca²⁺ responses to shear stress, ACh, and A-23187 (Figs. 2 and 3). ACh elicited a biphasic response followed by a plateau phase, whereas responses to shear stress or A-23187 developed much slower. It is likely that the rapid development of responses to ACh is due to the release of Ca²⁺ from intracellular stores because it was not prevented by removal of extracellular Ca²⁺ (Fig. 2D), whereas the secondary plateau phase is due most likely to an influx of Ca²⁺ via receptor-operated Ca²⁺ channels. The mechanisms of shear stress-induced increases of endothelial [Ca²⁺]_i suggested by previous studies include Ca²⁺ release from intracellular stores (14, 32) and Ca²⁺ entry from the extracellular medium (35). Because in the absence of extracellular Ca²⁺ shear stress-induced increases in endothelial [Ca²⁺]_i were abolished (Fig. 2D), it is likely that in the endothelium of intact skeletal muscle arterioles, shear stress stimulates only a minor influx of extracellular Ca²⁺ without activation of Ca²⁺ release from intracellular stores (35). We also confirmed that A-23187 stimulates receptor-independent entry of extracellular Ca²⁺ because its effect was abolished in a Ca²⁺-free medium (Fig. 3B). ACh- and histamine-induced Ca²⁺ responses were concentration dependent; however, maximal shear stress-induced Ca²⁺ response was reached at relatively low shear stress values and further increases in shear stress did not increase the magnitude of the response (Fig. 2, A and B). The findings that increasing concentration of A-23187, histamine, and ACh elicited substantial changes in endothelial [Ca²⁺]_i and NO-mediated dilations suggest (Fig. 3C) that elevation of endothelial [Ca²⁺]_i by these agonists corresponds to the activation of eNOS (10, 13, 26, 28). In contrast, increases in shear stress elicit NO-mediated arteriolar dilations similar in magnitude to those obtained with A-23187, histamine, or ACh without substantial increases in endothelial [Ca²⁺]_i (Fig. 3C), suggesting a significant role for Ca²⁺-independent activation of eNOS in the endothelial signal transduction of shear stress in skeletal muscle arterioles. It is of note that on the basis of the present findings, a role for increases in endothelial [Ca²⁺]_i in release of a non-NO, nonprostaglandin factor, such as endothelium-derived hyperpolarizing factor (EDHF), cannot be ruled out.

Recent studies have failed to demonstrate significant increases in endothelial [Ca²⁺]_i to shear stress in the intact rat aorta (18, 27) and hepatic portal vein (27), although shear stress induced a NO-mediated relaxation. In intact rabbit coronary arterioles, shear stress elicited, similar to the findings of the present study, only slight increases in endothelial [Ca²⁺]_i, but NO-mediated dilation was not abolished by chelating intracellular Ca²⁺ (31). Also, many studies of cultured endothelial cells showed that increases in shear stress do not lead to a substantial increase in [Ca²⁺]_i (1, 11, 38) unless ATP is present (7, 29). In isolated cremaster arterioles, increases in flow were shown to elicit increases in endothelial [Ca²⁺]_i (8). However, the function of endothelium in this vascular bed is known to significantly differ from that of skeletal muscle and coronary arterioles. For example, in cremaster arterioles, prostaglandins and not NO are the primary endothelial mediators, and stimulation of prostaglandin synthesis may require a greater increase in [Ca²⁺]_i than is required for NO synthesis.

Several mechanisms have been proposed to explain the Ca²⁺-independent release of endothelial mediators. Accordingly, various Ca²⁺-independent pathways for the mechanotransduction of shear stress and eNOS activation have been identified in endothelial cells (33), including changes in cytosolic pH (1, 11, 38), structure of the cytoskeleton (15), and activation of mitogen-activated protein (MAP) kinases (36) or the serine/tyrosine kinase Akt (9). Recently, it has been shown that in cultured endothelial cells, eNOS can be phosphorylated and stimulated by tyrosine kinases that can be activated by shear stress (1, 4, 9, 10, 13, 26, 28, 36).

Because in the present study we found that intraluminal application of the specific tyrosine kinase inhibitor genistein abolished shear stress-induced dilations (Fig. 4A) and because shear stress-induced NO-mediated dilations of coronary arterioles (which are independent of increases in endothelial [Ca²⁺]_i, Ref. 31) can also be blocked by genistein (30), it can be assumed that in skeletal muscle arterioles the shear stress-induced Ca²⁺-independent activation of eNOS involves phosphorylation by the tyrosine kinase pathway. Activation of the tyrosine kinase pathway is likely to be specific for the mechanotransduction of shear stress, because genistein affected neither ACh-induced arteriolar dilations nor the direct action of NO released from SNP on smooth muscle (Fig. 4B). The tyrosine kinase pathway may also link the mechanotransduction of shear stress to prostaglandin release because endothelial cyclooxygenase is known to be inhibited by genistein (34).

On the basis of our present and previous findings and the aforementioned studies, we developed a model for describing the different changes endothelial [Ca²⁺]_i and activation of eNOS in skeletal muscle arterioles by agonists and shear stress (Fig. 4C). Accordingly, it is likely that two distinct signaling pathways, primarily Ca²⁺-dependent and -independent pathways, are present in arteriolar endothelial cells. Vasoactive agents such as A-23187, histamine, and ACh elicit large increases in [Ca²⁺]_i due to an influx of extracellular Ca²⁺ and/or a simultaneous release of Ca²⁺ from intracellular stores and activate NO synthesis via a Ca²⁺-dependent pathway. ACh, but not A-23187, may also activate separate intracellular pathways that, together with increases in [Ca²⁺]_i, elicit the release of a non-NO, nonprostaglandin dilator factor. In contrast, increases in shear stress elicit NO synthesis and dilation of skeletal muscle arterioles by activation primarily of a Ca²⁺-independent pathway that involves eNOS activation by tyrosine kinases. Finally, activation of Ca²⁺-dependent and -independent pathways seems to depend on the nature of the physiological stimulus affecting arteriolar endothelial cells.