| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Department of Physiology, Monash University, Melbourne, Victoria, Australia
Submitted 17 January 2007 ; accepted in final form 24 April 2007
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
-nitro-L-arginine methyl ester (L-NAME, 50 mg/kg plus 10 mg·kg–1·h–1), to block nitric oxide (NO) production, altered basal hemodynamics, increasing mean arterial pressure (30 ± 3%) and reducing HBF (–30 ± 12%). Basal hemodynamics were not significantly altered by indomethacin (10 mg·kg–1·h–1), charybdotoxin (ChTx, 3 x 10–8 mol/l), apamin (2.5 x 10–7 mol/l), or ChTx plus apamin (to block endothelium-derived hyperpolarizing factor; EDHF). Hyperemic responses to local injection of acetylcholine (2.4 µg/kg) were reproducible in vehicle-treated mice and were not significantly attenuated by L-NAME alone, indomethacin alone, L-NAME plus indomethacin with or without co-infusion of diethlyamine NONOate to restore resting NO levels, ChTx alone, or apamin alone. Hyperemic responses evoked by acetylcholine were reduced by 29 ± 11% after combined treatment with apamin plus charybdotoxin, and the remainder was virtually abolished by additional treatment with L-NAME but not indomethacin. None of the treatments altered the hyperemic response to sodium nitroprusside (5 µg/kg). We conclude that endothelium-dependent vasodilation in the mouse hindlimb in vivo is mediated by both NO and EDHF. EDHF can fully compensate for the loss of NO, but this cannot be explained by tonic inhibition of EDHF by NO. Control of basal vasodilator tone in the mouse hindlimb is dominated by NO.
endothelium-derived hyperpolarizing factor; nitric oxide; prostanoids
Inhibitors of NO synthase (NOS) and cyclooxygenase (COX) have been available for in vivo use for many years. Approaches to inhibiting EDHF in vivo have been limited, due in large part to the fact that EDHF can involve different mechanisms according to the vascular bed and species, including diffusible factors such as epoxyeicosatrienoic acids, hydrogen peroxide, potassium, and direct electrical coupling between endothelial and vascular smooth muscle cells via myoendothelial gap junctions (8, 10, 30). One approach has been to target specific putative EDHFs. For example, in the renal circulation in vivo, a role for myoendothelial gap junctions has been implicated using connexin mimetic peptides (6), and an involvement of epoxyeicosatrienoic acids has been demonstrated using synthesis blockers (12, 22). In dog kidney in vivo, EDHF appears to be prominent in afferent juxtamedullary arterioles (17). In human forearm, EDHF-attributed increases in flow were markedly reduced by tetraethylammonium chloride, which blocks calcium-activated potassium channels (KCa) and voltage-sensitive K+ channels (13).
A consistent observation under in vitro conditions is that the EDHF-attributed response is blocked by a combination of charybdotoxin and apamin, which block big (B)-, intermediate (I)-, and small (S)-conductance KCa channels (4, 5, 7, 14). IKCa and SKCa channels are present predominantly on the endothelial cells, and blockade of these channels prevents the generation of hyperpolarization, which can spread to the underlying smooth muscle via myoendothelial gap junctions. These KCa channels also permit efflux of the diffusible EDHF, potassium. Functional BKCa channels on the smooth muscle are activated by epoxyeicosatrienoic acids to bring about vasodilatation. In the present study, we have taken advantage of the "universal" inhibiting properties of charybdotoxin and apamin against BKCa, IKCa, and SKCa channels and hence in blocking EDHF. In the hindlimb vascular beds of the rat in vivo, these agents all but abolish the hyperemic effects of acetylcholine that are resistant to inhibition of NOS and COX (20).
Our previous in vivo studies (20) and those of others (4, 17–19) employed the experimental paradigm standard in studying the biology of EDHF under in vitro conditions, where apamin and charybdotoxin are administered only after prior blockade of NOS and COX. Thus there is little information regarding the potential interactions between these endothelium-derived relaxing factors. Furthermore, inferences regarding the physiological roles of EDHF must be tempered with the caveat that EDHF has not been studied under conditions of intact NOS and COX. Therefore, in the present study, we set out to rectify this situation by investigating the relative roles of NO, vasodilator prostanoids, and EDHF in the control of hindlimb blood flow in vivo both in the presence and in the absence of the other vasodilators. Our experiments were performed in mice to open up the potential for studies in mice with targeted deletion of genes important in the biology of endothelium-dependent vasodilation (11, 26, 27).
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
Male C57BL/6J mice were obtained from Monash University Central Animal Services. All procedures were approved in advance by the Monash University Department of Physiology Animal Ethics Committee and were conducted in accordance with the Australian Code of Practice for the Care and Use of Animals for Scientific Purposes.
Surgical Preparation
At 12 wk of age, male C57BL/6J mice (25–31 g, n = 47) were anesthetized with isoflurane (Forthane, 4–5% vol/vol induction and 1.5–2.0% vol/vol maintenance; Abbott Australasia, Kurnell, NSW, Australia). Anesthesia was induced in a closed chamber and was maintained via a nose cone throughout surgery. Body temperature was maintained at 
37°C using a thermostatically controlled heating pad, fitted with a rectal thermometer. Arterial pressure was recorded via a catheter in the carotid artery. Saline (154 mmol/l NaCl) was continuously infused at 0.15 ml/h via the jugular vein throughout the entire experiment.
Hindlimb blood flow (HBF) was measured via a transit-time ultrasound flow probe (0.5 V, 0.7 mm; Transonic Systems) placed around the abdominal aorta, 2–3 mm proximal to the iliac bifurcation. A catheter was inserted in the left femoral artery and advanced to the iliac bifurcation to facilitate close arterial injection into the right hindlimb circulation. Saline solution was infused via this catheter throughout the entire experiment at 10 µl/min. The testicular and rectal vessels were ligated and the reproductive organs removed.
Basal mean arterial pressure (MAP), heart rate (HR), and HBF reached stable values within 5–10 min of completion of the preliminary surgical procedures. Hemodynamic measurements including HBF, systolic and diastolic arterial pressure, and calculated MAP and HR were then recorded digitally at 500 Hz and continuously displayed by a data acquisition program (Universal Acquisition, Univ. of Auckland, New Zealand) (23). During each experiment, data were saved continuously as 2-s averages of each variable. Hindlimb vascular conductance (HVC) was determined by dividing HBF by MAP.
General Experimental Protocol
All experiments followed a standard protocol. Bolus administration of the endothelium-dependent vasodilator acetylcholine was used to test endothelium-dependent vasodilator function in the hindlimb vasculature. Bolus administration of sodium nitroprusside was used to test endothelium-independent vasodilator function. Both agents were administered directly into the hindlimb circulation via the femoral artery catheter, in random order, at 5-min intervals. The bolus doses were administered in a volume of 10 µl over 10 s followed by prompt resumption of saline infusion at 10 µl/min. In each study, before commencing the experiment proper, we tested responses to a range of doses of acetylcholine (1.6, 2.4, and 4 µg/kg) and sodium nitroprusside (3.2, 5, and 6 µg/kg) in each mouse. Invariably, we found that the lowest dose tested had little effect on HBF, whereas the highest dose resulted in appreciable falls in MAP. Therefore, in the experiment proper we chose to administer only single doses of acetylcholine (2.4 µg/kg) and sodium nitroprusside (5 µg/kg).
Responses to acetylcholine and sodium nitroprusside were tested three times in all mice. In each mouse, the initial responses to acetylcholine and sodium nitroprusside were defined as the control responses (period 1). Specific treatments were then administered before responses to acetylcholine and sodium nitroprusside were retested (period 2). Responses to acetylcholine and sodium nitroprusside were then tested for the third time after administration of an additional treatment (period 3).
Pharmacological agents were administered to inhibit various components of endothelium-dependent vasodilation. To inhibit NOS, N-nitro-L-arginine methyl ester (L-NAME; Sigma Chemical, St. Louis, MO) was administered intravenously as a bolus of 50 mg/kg followed by an infusion of 10 mg·kg–1·h–1. To inhibit COX, indomethacin (Sigma) was administered as an intravenous infusion of 10 mg·kg–1·h–1. An equilibration period of at least 20 min was allowed, after treatment with indomethacin and/or L-NAME commenced, before responses to acetylcholine and sodium nitroprusside were retested. Under conditions of NOS inhibition, vasodilator tone was restored using the NO donor diethlyamine NONOate (DEA/NO; Alexis Biochemicals, Lausen, Switzerland). This was administered intravenously at a dose of 25 µg·kg–1·min–1, titrated to restore MAP to its levels before L-NAME administration. Once infusions of indomethacin, L-NAME, and/or DEA/NO commenced, they continued for the remainder of the experiment. L-NAME and indomethacin were administered by the intravenous route rather than by local infusion, since the 30 min required to ensure complete blockade of NO and prostanoid synthesis would likely also result in significant spillover into the systemic circulation. For consistency, the NO "replacement," DEA/NO, was also administered systemically.
Charybdotoxin and apamin (Auspep, Parkville, VIC, Australia) were used to block the potassium channels mediating EDHF. This approach blocked all known channels implicated in EDHF and allowed comparisons with the numerous studies in which these agents have been used to block EDHF in blood vessels from the mouse under in vitro conditions (4, 7, 14). At the concentrations used, the blockade by charybdotoxin and apamin was complete within 10 min, and the effects were rapidly reversible. We administered these agents directly into the hindlimb in a volume of 10 µl/min to minimize the possible systemic effects of these toxins. The concentration of these agents was adjusted, on the basis of HBF in each individual mouse, to achieve a blood concentration in the hindlimb of 30 nmol/l for charybdotoxin and 250 nmol/l for apamin. An equilibration period of at least 10 min was allowed, after commencing infusions of charybdotoxin and/or apamin, before responses to acetylcholine and sodium nitroprusside were retested. When charybdotoxin and/or apamin were administered during period 2, the infusions ceased after responses to acetylcholine and sodium nitroprusside had been tested.
All of the drugs used were dissolved in 154 mmol/l NaCl (saline) except indomethacin stock solution (0.1 mol/l), which was dissolved in 0.1 M Na2CO3 and was subsequently diluted in saline.
Specific Experimental Protocols
Mice were randomized to one of seven groups. In each group, different combinations of the various inhibitors were administered to determine the relative contributions of various endothelium-derived factors to acetylcholine-induced hyperemia in the hindlimb, and to confirm that these factors do not influence sodium nitroprusside-induced hindlimb hyperemia.
Group 1: time controls and effects of vehicle treatment (n = 6). The stability of the responses to acetylcholine and sodium nitroprusside across the time course of the experiment was tested. Once control responses to acetylcholine and sodium nitroprusside were tested (period 1), a solution of 154 mmol/l NaCl was administered intravenously as a bolus (0.33 ml/kg) followed by an infusion (0.1 ml·kg–1·min–1) that continued for the remainder of the experiment. This treatment was the vehicle for administration of L-NAME and indomethacin (see group 2). Twenty minutes after the intravenous saline bolus, responses to acetylcholine and sodium nitroprusside were retested (period 2). After hemodynamic variables had stabilized and an additional 20-min period had elapsed, responses to acetylcholine and sodium nitroprusside were tested for the third time (period 3).
Group 2: effects of indomethacin and the superimposition of L-NAME (n = 6). The effects of progressive inhibition of COX and NOS were examined. Responses to acetylcholine and sodium nitroprusside were tested under control conditions (period 1), 20 min after commencement of treatment with indomethacin (period 2), and then 20 min after commencement of further treatment with L-NAME (period 3).
Group 3: effects of apamin and the superimposition of charybdotoxin (n = 6). The effects of progressive blockade of SKCa channels and then BKCa and IKCa channels were examined. Responses to acetylcholine and sodium nitroprusside were tested under control conditions (period 1), 10 min after commencement of apamin infusion (period 2), and 10 min after commencement of an infusion containing charybdotoxin plus apamin (period 3).
Group 4: effects of charybdotoxin and the superimposition of apamin (n = 5). The effects of progressive blockade of BKCa and IKCa channels and then SKCa channels were explored. Responses to acetylcholine and sodium nitroprusside were tested under control conditions (period 1), 10 min after commencement of charybdotoxin infusion (period 2), and 10 min after commencement of infusion of apamin plus charybdotoxin (period 3).
Group 5: effects of charybdotoxin plus apamin and the superimposition of L-NAME plus indomethacin (n = 5). The effects of combined blockade of KCa channels were examined under conditions of intact COX and NOS, and then after additional inhibition of COX and NOS. Responses to acetylcholine and sodium nitroprusside were tested under control conditions (period 1) and 10 min after commencement of treatment with charybdotoxin plus apamin (period 2). Treatment with L-NAME plus indomethacin then commenced, and 10 min later, the infusion of charybdotoxin plus apamin recommenced. Responses to acetylcholine and sodium nitroprusside were retested 10 min after the infusion of charybdotoxin plus apamin recommenced (period 3).
Group 6: effects of L-NAME plus indomethacin, and the superimposition of charybdotoxin plus apamin (n = 6). We examined the effects of combined inhibition of NOS and COX, and then additional inhibition of KCa channels. Responses to acetylcholine and sodium nitroprusside were tested under control conditions (period 1), 20 min after commencement of treatment with L-NAME plus indomethacin (period 2), and then 10 min after commencement of an infusion of charybdotoxin plus apamin (period 3).
Group 7: effects of DEA/NO under conditions of COX and NOS inhibition, and the superimposition of charybdotoxin plus apamin (n = 7). We examined the effects of restoring NO levels under conditions of NOS and COX inhibition, and then on additional blockade of KCa channels. Responses to acetylcholine and sodium nitroprusside were tested 20 min after commencement of treatment with L-NAME and indomethacin (period 1), 20 min after commencement of treatment with DEA/NO (period 2), and 10 min after further treatment with charybdotoxin plus apamin (period 3).
Supplementary experiments. In an additional group of six mice, we tested the effects of inhibition of NOS or COX alone during blockade of KCa channels. Responses to acetylcholine and sodium nitroprusside were tested under control conditions (period 1). Treatment with either L-NAME (3 mice) or indomethacin (3 mice) then commenced, and 10 min later, the local infusion of charybdotoxin plus apamin commenced. Responses to acetylcholine and sodium nitroprusside were retested 10 min after the infusion of charybdotoxin plus apamin commenced (period 2).
Data Analysis and Statistical Methods
Basal levels of MAP, HR, HBF, and HVC were defined as the average levels during the 2-min period before responses to acetylcholine and sodium nitroprusside were tested during each experimental period. Responses to sodium nitroprusside and acetylcholine were analyzed both in terms of the peak (maximal) changes in HBF, HVC, MAP, and HR and as the integrated responses (area under the curve). The basal variables and responses to acetylcholine and sodium nitroprusside in periods 1, 2, and 3 for each individual animal were analyzed as paired data using ANOVA and Tukey's post hoc test (16) to determine whether these differed among the three experimental periods in each group. Data are expressed as means ± SE. All statistical tests were performed using the software package SYSTAT (version 11; SPSS, Chicago, IL). P 
0.05 was accepted as statistically significant.
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
In group 1 (time controls), in which only vehicle treatments (infusion and bolus administration of saline) were administered, basal MAP and HR were relatively stable across the course of the experiment. HBF gradually increased across the course of the experiment, so that it was 24 ± 4% greater during period 3 than during the control period (Table 1). HVC also increased across the course of the experiment in this control group. Thus time-dependent changes in HBF and HVC must be taken into account when interpreting observations in the groups receiving active treatments. In group 2, administration of indomethacin alone did not significantly alter any of the hemodynamic variables, but subsequent administration of L-NAME significantly increased MAP (24 ± 5% compared with period 2) and decreased HBF (27 ± 6% compared with period 2). These effects had stabilized by 20 min after L-NAME treatment commenced. In group 3, apamin treatment was associated with a small but statistically significant increase in HVC (16 ± 5%), but no other significant changes in basal hemodynamics. The apparent effect of apamin on HVC may simply reflect the gradual drift in this parameter that we observed in group 1. Subsequent combined treatment with apamin plus charybdotoxin was associated with a small increase in MAP (9 ± 1% compared with period 2), while HVC returned to a level close to control. When the order of administration of apamin and charybdotoxin was reversed in group 4, charybdotoxin alone did not significantly alter basal hemodynamic variables. The addition of apamin to the charybdotoxin treatment was associated with a significant increase in HBF (22 ± 8% compared with control) but no significant change in MAP. As for group 3, we cannot be certain that this increase in HBF is attributable to the actions of apamin, since this variable also gradually increased across the course of the experiment in vehicle-treated mice (group 1). In group 5, co-administration of charybdotoxin plus apamin had no marked effects on MAP or HBF, but further addition of L-NAME plus indomethacin was associated with increased MAP and reduced HVC, as was observed when L-NAME was administered on a background of indomethacin infusion, without apamin and charybdotoxin, in group 1. In group 6, L-NAME plus indomethacin treatment was associated with increased MAP (30 ± 3%) and reduced HBF (30 ± 12%). Subsequent administration of apamin plus charybdotoxin did not significantly alter basal hemodynamic variables. In group 7, on a background of prior inhibition of NOS and COX, DEA/NO significantly reduced MAP (25 ± 3%) and increased HBF (115 ± 14%), so that these variables did not differ significantly from their levels before administration of L-NAME plus indomethacin. Subsequent further treatment with charybdotoxin plus apamin did not significantly alter basal hemodynamic variables (Table 1).
|
Close arterial injection of acetylcholine induced transient increases in HBF and HVC (Fig. 1) that commenced within 1–2 s of the injection and had fully recovered by 200 s after the injection. When averaged over all 41 mice from groups 1–7, the peak increase in HBF induced by acetylcholine was 0.77 ± 0.03 ml/min. The hyperemic response to acetylcholine was accompanied by a small but significant depressor response (–5 ± 1 mmHg maximum). The combined effects of increased HBF and decreased MAP resulted in an overall increase in HVC (peak = 11.38 ± 0.05 µl·min–1·mmHg–1). In subsequent analyses, responses to acetylcholine are characterized as the area under the curve, to take into account both the amplitude and duration of the vasodilator response.
|
0.001; Figs. 6 and 7). In group 7, when NO was restored after L-NAME administration by co-infusion of DEA/NO, acetylcholine-induced hyperemia was not different from that observed following L-NAME/indomethacin treatment alone (Fig. 8). As was observed in groups 5 and 6 in the absence of DEA/NO, during DEA/NO infusion in group 7, the hyperemic response to acetylcholine was only reduced by additional treatment with charybdotoxin and apamin (area under the HBF curve reduced by 78 ± 18%, P 0.001; Fig. 8).
|
|
|
|
|
|
|
|
Responses to Sodium Nitroprusside
Bolus administration of the endothelium-independent vasodilator sodium nitroprusside produced transient increases in HBF and HVC (Fig. 1). The hyperemic response was apparent within 1–2 s of the injection, and HBF had fully recovered by 160 s after the injection. This resulted in maximum increases in HBF and HVC of 0.42 ± 0.05 ml/min and 5.31 ± 0.69 µl·min–1·mmHg–1, respectively (average of control responses in the 41 mice in groups 1–7). There were few or no changes in MAP and HR (–3 ± 1 mmHg and –1 ± 2 beats/min, respectively). As with responses to acetylcholine, the overall responses to sodium nitroprusside were characterized as the area under the curve. The integrated responses of MAP, HBF, and HVC to sodium nitroprusside were stable across the course of the experiment in vehicle-treated mice (group 1; Fig. 2) and in all groups of mice receiving active treatments (groups 2–7; Figs. 3–8).
Responses to Vehicle (Saline)
Bolus administration of the saline vehicle (10 µl) directly into the femoral artery had no appreciable effects on HBF, MAP, HVC, or HR (Fig. 1).
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
29% by blockade of the KCa channels thought to mediate EDHF. Thus our data provide evidence that EDHF contributes to endothelium-dependent vasodilatation in the mouse hindlimb circulation in vivo. Considered in isolation, these data could also be interpreted as evidence against roles for NO and vasodilator prostanoids in acetylcholine-stimulated, endothelium-dependent vasodilation in the mouse hindlimb circulation. However, combined blockade of NOS and of KCa channels, with or without COX inhibition, virtually abolished acetylcholine-induced hyperemia in this vascular bed. Collectively, these observations suggest that both NO and EDHF mediate endothelium-dependent vasodilation in the mouse hindlimb in vivo, but that EDHF can fully compensate for the absence of NO. Thus our study extends the observations of Brandes et al. (4) who showed that prior blockade of either NO or EDHF was required to unveil the contributing nature of both endothelium-dependent vasodilator systems in mouse hindlimb in vitro. Our second major finding was that, under conditions of combined inhibition of NOS and COX, restoration of NO by infusion of DEA/NO did not significantly alter the hyperemic response to acetylcholine. This observation suggests that the ability of EDHF to compensate for the absence of NO cannot be explained by tonic inhibition of EDHF by NO.
After combined treatment with L-NAME and indomethacin, acetylcholine-induced hyperemia was virtually abolished by combined inhibition of KCa channels with apamin and charybdotoxin. Thus the vasodilator response to acetylcholine that remains after inhibition of COX and NOS can be attributed to EDHF. Combined treatment with apamin plus charybdotoxin blunted acetylcholine-induced hyperemia, but we could only detect this effect when we pooled data from all 16 mice that received this treatment. Nevertheless, these data suggest that EDHF makes an important contribution to endothelium-dependent vasodilatation in the mouse hindlimb in vivo. In mice treated with charybdotoxin and apamin, acetylcholine-induced hyperemia was virtually abolished by L-NAME plus indomethacin or L-NAME alone but not by indomethacin alone. Thus the hyperemic response to acetylcholine that remains after blockade of the KCa channels that mediate EDHF can be attributed to NO.
What mechanisms allow EDHF to compensate for the absence of NO? NO can inhibit the EDHF response under some experimental conditions. For example, NO donors inhibited EDHF-mediated dilation in rabbit carotid arteries and porcine epicardial arteries, and EDHF-mediated hyperpolarization of cultured rat aortic smooth muscle cells in vitro (2). These effects appear to be mediated by inhibition of EDHF release, perhaps through inhibition of calcium signaling in endothelial cells and/or inhibition/downregulation of EDHF-generating enzymes (2, 31). Similar mechanisms may operate within the renal circulation in vivo, since renal hyperemic responses to acetylcholine and bradykinin in anesthetized rabbits were not blunted by N-nitro-L-arginine (L-NNA) alone but were markedly attenuated by L-NNA when NO was restored by infusion of glyceryl trinitrate (21), suggesting suppression of EDHF by NO. A similar phenomenon has been observed in the canine coronary circulation in vivo (19). Also, endothelium-dependent vasodilation persists in skeletal muscle arteries from mice with targeted deletion of endothelial NOS (eNOS) (11) and in mesenteric arteries of mice with targeted deletion of both eNOS and COX-1 (26). The results of studies using pharmacological tools indicate that these relaxations were mediated by EDHF. To determine whether an inhibitory effect of NO on EDHF underlies the ability of EDHF to compensate for the loss of NO in endothelium-dependent vasodilation in the mouse hindlimb, we tested the effects of restoring NO after NOS and COX blockade by infusing the NO donor DEA/NO. The dose of DEA/NO was titrated to restore MAP to its control level before treatment with L-NAME and indomethacin. This treatment did not significantly alter the hyperemic response to acetylcholine, which was in turn virtually abolished by charybdotoxin plus apamin, indicating that NO does not inhibit EDHF-mediated vasodilation in the mouse hindlimb circulation in vivo. NO also appears not to inhibit EDHF-mediated vasodilation in the human forearm circulation in vivo (25) or in rat middle cerebral arteries in vitro (24). Thus, although NO-mediated inhibition of EDHF likely explains the ability of EDHF to compensate for the absence of NO in some experimental systems, other mechanisms likely operate within the mouse hindlimb circulation in vivo.
So what might these mechanisms be? One possibility is that the signaling cascades for NO and EDHF converge. At high concentrations, NO can hyperpolarize vascular smooth muscle (29) at least partly through activation of charybdotoxin-sensitive KCa channels (1, 3). Alternatively, there may be some kind of ceiling effect associated with both NO and EDHF action, so that despite the fact that both systems operate simultaneously in the mouse hindlimb circulation, activation of EDHF is capable of inducing maximal vasodilation.
Our present observations shed new light on the physiological function of EDHF. EDHF has been considered as a "backup" system that comes into play mainly under pathological conditions of reduced NO bioavailability and has minimal function under NO-replete conditions (19). This view has arisen in large part from the evidence that NO can inhibit EDHF function (2, 19, 31) and that blockade of putative EDHF pathways alone often does not blunt endothelium-dependent vasodilation in the absence of endothelial dysfunction (4, 9, 28). Our present findings suggest that this is an oversimplistic interpretation; if the same logic were applied to our present observations, one might conclude that NO does not play much of a role in agonist-evoked vasodilation under EDHF-replete conditions. We believe that our present observations indicate that NO and EDHF operate as parallel pathways in acetylcholine-induced vasodilation in the mouse hindlimb. However, the complex interactions between NO and EDHF preclude quantification of their relative contributions to endothelium-dependent vasodilation.
Basal HBF was not significantly reduced by apamin and/or charybdotoxin, regardless of whether these agents were administered alone or in combination or under conditions of intact or inhibited NOS and COX. These observations are consistent with our previous findings in both the hindlimb and mesenteric circulations of rat (20) and suggest that EDHF plays at most a small role in the control of basal tone in these vascular beds. Si et al. (27) recently demonstrated that mice lacking a functional gene for the intermediate-conductance KCa channel are hypertensive. Their observations are not necessarily at odds with ours, since acute, local (hindlimb) blockade of the intermediate-conductance KCa channel hardly equates to the life-long and widespread absence of this protein. There are also reports of vasoconstriction in the isolated perfused hindlimbs of both rats (15) and mice (4) induced by charybdotoxin and apamin. These preparations were perfused with physiological saline solution, not blood as in our in vivo study, and charybdotoxin commonly evokes constriction in isolated saline superfused arterial segments (5).
Our present observations in vivo are in large part consistent with those of previous investigations of the mouse hindlimb vasculature in vitro. For example, Brandes et al. (4) found that blockade of NOS and COX in the mouse isolated perfused hindlimb did not attenuate acetylcholine-induced vasodilation, while subsequent treatment with charybdotoxin and apamin or 40 mmol/l potassium virtually abolished acetylcholine-induced vasodilation (4). Acetylcholine-induced hindlimb vasodilation was not affected by inhibitors of cytochrome P450-dependent arachidonate metabolism but was greatly inhibited by CB1 cannabinoid receptor agonists and gap junction uncouplers, although only under conditions of NOS blockade (4). Collectively, these data suggest that gap junctions and KCa channels are critical for full expression of EDHF effects in mouse hindlimb vasculature, and that the dominant EDHF in mouse hindlimb is not a product of arachidonate. The situation in the rat hindlimb circulation is somewhat different, since carbachol-induced vasodilation resistant to NOS blockade can be blunted by cytochrome P450 inhibitors (15). Interestingly, unlike the situation in mouse hindlimb, EDHF cannot fully compensate for the loss of NO in rat hindlimb, since NOS inhibition alone reduced the amplitude of the hyperemic response to acetylcholine (20). Thus there appear to be species differences in both the identity and function of EDHF in the hindlimb circulations of rats and mice.
A limitation of our present study must be considered. HBF and HVC gradually increased across the course of the experiment in mice receiving only vehicle treatments (group 1). This phenomenon reduced our ability to detect changes in basal HBF and HVC. Consequently, the small increases in these variables observed after treatment with apamin in groups 3 and 4 may not be specific effects of blockade of small-conductance KCa but rather nonspecific time-related changes in the experimental preparation.
In summary, our present observations suggest that both NO and EDHF contribute to endothelium-dependent vasodilation in the mouse hindlimb in vivo. EDHF can fully compensate for the loss of NO, but this appears not to be due to inhibition of EDHF by NO. Because NO and EDHF appear to operate as parallel pathways, it is not possible to quantify their relative contributions to endothelium-dependent vasodilation. Nevertheless, our present findings challenge the view that EDHF is merely a backup system that comes into play under conditions of reduced NO bioavailability. Rather, in mouse hindlimb, both systems appear to operate simultaneously in a nonadditive fashion.
|
GRANTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  A. J. Edgley, M. Tare, R. G. Evans, C. Skordilis, and H. C. Parkington In vivo regulation of endothelium-dependent vasodilation in the rat renal circulation and the effect of streptozotocin-induced diabetes Am J Physiol Regulatory Integrative Comp Physiol, September 1, 2008; 295(3): R829 - R839. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. L. Gaubert, D. Sigaudo-Roussel, M. Tartas, G. Berrut, J. L. Saumet, and B. Fromy Endothelium-derived hyperpolarizing factor as an in vivo back-up mechanism in the cutaneous microcirculation in old mice J. Physiol., December 1, 2007; 585(2): 617 - 626. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
P 
