Flavin adenine dinucleotide (FAD) facilitates electron transfer processes in flavoproteins.1,2 FAD exists in most cells and in nmol/L concentrations in plasma,3 is released by cardiomyocytes during inflammation,4 and protects against ischemia-reperfusion injury to the rat myocardium by extracellular rather than intracellular mechanisms.5 The finding that FAD elicits an endothelium-dependent vasodilation in isolated perfused rat mesenteric beds by activation of endothelial P2Y-purinoceptors,6 suggests that FAD releases endothelium-derived relaxing (EDRF) and/or hyperpolarizing (EDHF) factors.7 However, the dose-dependent vasodilator responses elicited by systemic injections of FAD in anesthetized rats were not attenuated by high doses of the P2Y/P2X-purinoceptor antagonist, suramin, the nitric oxide synthase (NO) synthase (NOS) inhibitor, NG-nitro-L-arginine (L-NAME), or the prostaglandin synthesis inhibitor, indomethacin.8 This suggests that FAD may release an EDRF/EDHF other than NO or prostacyclin. Moreover, the observation that flavin mononucleotide (FMN) elicited negligible hemodynamic responses,8 raises the possibility that FAD may be an agonist at recently identified adenine receptors.9-11
There is compelling evidence that the S-nitrosothiol, L-S-nitrosocysteine (L-SNC), is an EDRF,12,13 an EDHF,14,15 and a neurotransmitter/neuromodulator.16 Although the de novo synthesis and release of S-nitrosothiols may occur upon Ca2+-dependent activation of NOS in these cells,17 there is substantial direct evidence that preformed pools of S-nitrosothiols or dinitrosyl-iron complexes exist in a variety of mammalian tissues, including the vasculature,18,19 brain,20 and acinar21 and red blood22 cells. Moreover, there is considerable in vivo and in vitro evidence that vascular endothelial cells23-25 and lumbar autonomic neurons26,27 release preformed pools of nitrosyl factors and that vascular smooth muscle,28 gastric fundus,29 and corpus cavernosum30 contain photosensitive stores of nitrosyl factors. Moreover, pancreatic acinar cells and dorsal root ganglion cells contain pools of heavy-molecular weight molecular S-nitrosothiols that are induced to release NO within the cells via Ca2+-dependent mechanisms.21 The first injection of higher doses of the endothelium-dependent agonists, acetylcholine and bradykinin, elicited robust vasodilator responses in L-NAME-treated conscious rats, whereas successive injections elicited progressively smaller responses.23 Since the vasodilator actions of L-SNC and NO-donors such as sodium nitroprusside (SNP) were augmented in these rats,23 the progressive loss of response to acetylcholine and bradykinin was consistent with “use-dependent” depletion of preformed pools of endothelial nitrosyl factors (eg, L-SNC), which could not be replenished in the absence of NO synthesis. This use-dependent loss of endothelium-dependent vasodilation also occurs in vitro.24,25
The aim of this study was to provide evidence that intravenous injections of a 2.5 μmol/kg dose of FAD (FAD2.5) may release preformed pools of endothelial nitrosyl factors in conscious rats. In these studies, the vasodilator responses elicited by 6 successive injections of FAD2.5 were determined in saline-treated or L-NAME-treated rats. To better determine whether a loss of response to FAD2.5 in L-NAME-treated rats involves depletion of preformed pools of nitrosyl factors, the responses elicited by acetylcholine and bradykinin, L-SNC and SNP were also determined. The muscarinic M3 receptor is a major sub-type of cholinergic receptor on vascular endothelial cells.31-34 We anticipated that the vasodilator responses elicited by acetylcholine would be diminished in L-NAME-treated rats that had received multiple injections of FAD2.5 due to depletion of preformed pools of nitrosyl factors. To determine that the potential loss of response to acetylcholine in L-NAME-treated rats was not due to downregulation of muscarinic M3 receptors, we determined the affinities and densities of these receptors in endothelial membranes from thoracic aortae of the various treatment groups using the selective M3 receptor ligand, 3H-4-diphenylacetoxy-N-methylpiperidine methiodide (3H-4DAMP).35,36
All studies were carried out in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals (NIH Publications No. 80-23) revised in 1996. The protocols were approved by the Animal Care and Use Committee of the University of Georgia. Male Sprague-Dawley rats (Harlan, Madison, WI) weighing 268 ± 6 g were used in these studies.
Surgical Procedures and Description of Study Groups
The rats were anesthetized with sodium pentobarbital (50 mg/kg, IP), and a catheter (PE-50) was implanted into the left carotid artery to measure pulsatile and mean arterial blood pressures (MAP). A catheter (PE-50) was placed in the right jugular vein to inject drugs. The left lumbar sympathetic chain was isolated, cut, and removed caudally to the bifurcation of the left common iliac artery and vein, as described previously.23 The right sympathetic chain was left intact. A midline laparotomy was performed, and a miniature pulsed Doppler flow probe was placed on the superior mesenteric artery to continuously record mesenteric blood flow velocity and to determine mesenteric vascular resistances (MR). Doppler flow probes were also placed on the left and right common iliac arteries to record hindlimb blood flow velocities and to determine vascular resistance in the sympathetically denervated hindlimb (HLRD) and sympathetically intact hindlimb (HLRI), respectively.23,37,38 In some of the rats, a Doppler flow probe was placed around the ascending aorta to continuously record cardiac output and to calculate total peripheral resistances (TPR).39,40 These rats were used to define the hemodynamic actions of the 0.25 and 2.5 μmol/kg doses of FAD as fully as possible. Since the major aim of the present study was to examine the possibility that FAD elicits the release of endothelial stores of S-nitrosothiols, the rest of the experiments determined the vasodilator and/or vasoconstrictor actions of FAD in the mesenteric and hindlimb (intact and sympathetically-denervated) vascular beds.
The probes were sutured in place, and the leads and catheters were tunneled subcutaneously and exteriorized between the scapulae. To protect the probe wires and polyethylene tubing while allowing the rats unrestricted movement during recovery and experimental testing, the free ends of the catheters and Doppler leads were led through a stainless steel skin button spring swivel assembly that was mounted to a ring stand clamp and suspended above the cage. All wounds were liberally coated with triple antibiotic (neomycin, polymixin B, and bacitracin) ointment (Fougera, NY) and then sutured. The rats were returned to their home cages in a room maintained on a 12-h light/dark cycle and were allowed 5 days to recover from the surgeries. Food and water were freely available. Details of the Doppler technique, including construction of the probes, reliability of recording flow velocity (Doppler shift, kHz) and determination of regional vascular resistances have been described by Haywood et al.41 Regional vascular resistances and TPR at any point in time were determined by the following formula: resistance = MAP/blood flow velocity (mm Hg/kHz Doppler shift). There were 10 rats in each group in the in vivo studies. The numbers of rats used for the ligand-binding studies will be described below.
In Vivo Protocols
On the day of the experiments, the arterial catheter was connected to a Beckman dynagraph-coupled pressure transducer to record MAP. The Doppler leads were connected to a Beckman Dynagraph-coupled Doppler Flowmeter (Bioengineering, University of Iowa) to record blood flows. After allowing the rats 30 min to settle, the hemodynamic responses elicited by intravenous injections of the test agents were determined.
Group 1-4 rats received bolus injections of acetylcholine (1.0 μg/kg), L-SNC (100 nmol/kg), SNP (8 μg/kg), bradykinin (5 μg/kg), and FAD (0.25 μmol/kg, FAD0.25) in a latin-square design (see Table 1). Two rats in each group received the injections of acetylcholine, L-SNC, SNP, BK, and then FAD0.25 (eg, see Group 1), whereas the orders of administration in the remaining 4 pairs of rats are shown in the columns designated Groups 2-5. Each injection was given 10 min apart to allow the responses to subside completely before another injection was given. Thirty minutes after the last injection of the above agents, Groups 1 and 2 received a single bolus injection of saline; after 15 minutes; the rats received 6 consecutive injections of saline (Group 1, 1saline to 6saline) or a 2.5 μmol/kg dose of FAD (Group 2, 1FAD2.5 to 6FAD2.5), given 5 min apart. Groups 3 and 4 received a single bolus injection of L-NAME (50 μmol/kg); after 15 minutes, the rats received 6 consecutive injections of saline (Group 3) or FAD2.5 (Group 4), as above. Group 5 rats received a bolus injection of L-NAME (50 μmol/kg); after 1 minute, an injection of indomethacin (10 mg/kg),8 and after 15 min the rats received 6 consecutive injections of FAD2.5, as above. Ten min after the 6th injection of saline or FAD2.5, injections of acetylcholine, L-SNC, SNP, bradykinin and FAD0.25 were given 10 min apart. Each drug injectate (2-20 μL) was flushed into the rat by an 80 μL injection of saline.
Preparation of Endothelial Cell Membranes
Endothelial membranes from rat thoracic aortae were prepared for receptor binding as described by Cheng et al.36 Four groups of conscious rats were prepared with femoral artery and venous catheters. Groups 6 and 7 (n = 16 per group) received an injection of saline and then 6 injections of saline or FAD2.5, respectively. Groups 8 and 9 (n = 16 per group) received an injection of L-NAME (50 μmol/kg). Ten min after the last injection of saline or FAD2.5, the rats were decapitated, and the thoracic aortae from each group were placed in ice-cold Tris/ethylenediamine-tetraacetic acid (EDTA) buffer containing 50 mM Tris HCl, 5 mM EDTA, and 0.1 mM phenyl-methyl-sulfonyl-flouride (PMSF, pH 7.4). The arteries were stripped clean of fat and connective tissue and cut into 2 segments of about 1.5 cm. Each segment was everted over tapered glass pipettes and immersed in 5 mL of the above ice-cold buffer for 15 min. To selectively rupture and remove the endothelium from the aortae, the preparations were sonicated for 25 min and vortexed for 5 min. Using scanning electron microscopy, Cheng et al36 demonstrated that the endothelium but not the basement membrane was removed during these steps. The cloudy solution of ruptured endothelial cells was centrifuged at 45,000 × g for 45 min and washed with buffer. The final pellet was resuspended in 275 μL of buffer, and 25 μL was taken to determine protein concentrations using bovine serum albumin as standard.42 The studies were repeated another 2 times in arteries from different groups of rats, and the means (± SEM) of the 3 sets of results were determined.
Radioligand Binding Studies
Receptor binding assays with 3H-4DAMP were adapted from Cheng et al36 and Michel et al.43 In the saturation experiments, aliquots of endothelial membrane preparations (40 μg of protein) were incubated in assay buffer (50 mM Tris-HCl, pH 7.4) with 0.1-25 nM concentrations of 3H-4DAMP (specific activity, 75.4 Ci/mmol; Du Pont NEN) in a final volume of 400 μL. After incubation for 45 min at room temperature, the bound and free 3H-4DAMP were separated by vacuum filtration through Whatman GF/B glass fiber filters that were pretreated with 0.2% polyethylene-imine. The filter disks were washed 3 times with ice-cold assay buffer and counted for trapped radioactivity. Nonspecific binding was determined by adding 500 times excess of cold 4DAMP. Dissociation constants (KD) and maximal binding (Bmax) were calculated from the Scatchard plots by nonlinear regression analyses.44
All drugs were obtained from Sigma (St. Louis, MO) except for pentobarbital sodium and sodium nitroprusside, which were obtained from Abbott (Chicago, IL). All drugs were dissolved and diluted for injection in sterile saline except for indomethacin, which was dissolved in dimethylsulfoxide prior to dilution in saline immediately before injection.8 L-SNC was synthesized immediately before use as detailed previously.37,38
All data are presented as mean ± SEM. The in vivo data were analyzed by repeated-measures analysis of variance (ANOVA)45 followed by Student modified t-test with Bonferroni corrections for multiple comparisons between means.46 The receptor binding data were analyzed by 1-way ANOVA45 followed by Student modified t-test with the Bonferroni correction for multiple comparisons between means.46 A value of P < 0.05 was taken to denote statistical significance.
Hemodynamic Responses Elicited by FAD
The hemodynamic responses elicited by the 0.25 and 2.5 μmol/kg doses of FAD (FAD0.25 and FAD2.5, respectively) are summarized in Figure 1. FAD0.25 elicited a decrease in MAP that was initially associated with a decrease in cardiac output and decreases in TPR, HLRI, and HLRD, whereas there was an increase in MR (Phase 1 responses). Within 15 sec, the reduction in MAP was associated with pronounced decreases in TPR, HLRI, HLRD, and MR, whereas there was an increase in cardiac output (Phase 2 responses). Taken together, it is apparent that the Phase 1 decrease in MAP was due to decreases in cardiac output and regional vascular resistances, whereas the Phase 2 reduction in MAP was due only to decreases in vascular resistances. The injections of FAD2.5 elicited a similar pattern of responses except that Phase 1 increases in MR were not observed. The Phase 1 and 2 reductions in MAP, CO, TPR, and regional vascular resistances elicited by FAD2.5 were substantially greater and of longer duration than those elicited by FAD0.25.
The injection of 100 μL of saline (equivalent to the maximal volume of FAD injectate) elicited negligible hemodynamic responses. For example, the changes in heart rate and MAP elicited by bolus injections of saline in naïve rats were +1 ± 1% and 0 ± 0%, respectively (P > 0.05, for both responses). In addition, the changes in heart rate and MAP elicited by bolus injections of saline in L-NAME-treated rats were −1 ± 1% and 1 ± 1%, respectively (P > 0.05, for both responses).
Phase 1 Responses Elicited by FAD2.5 in Saline-Treated or L-NAME-Treated Rats
The Phase 1 responses elicited by injections 1-6 (inj1-6) of FAD2.5 after injection of saline or L-NAME are summarized in Figure 2. Inj1 of FAD2.5 elicited pronounced decreases in MAP, HLRI, and HLRD, but it did not affect MR in saline-treated rats. Inj2-6 of FAD2.5 elicited similar responses except that the decreases in HLRD diminished slightly upon repeated injection (asterisks in Figure 2 denote significance based on arithmetic differences between inj5-6 and inj1 responses). Inj1 of FAD2.5 in L-NAME-treated rats elicited similar responses to those of inj1 in saline-treated rats. Inj2-6 FAD2.5 elicited progressively smaller decreases in MAP, HLRI, and HLRD. The loss of vasodilation in the denervated hindlimb occurred more rapidly than in the intact hindlimb. Moreover, addition, inj3-6 of FAD2.5 elicited progressively greater increases in MR in the L-NAME-treated rats.
Phase 2 Responses Elicited by FAD2.5 in Saline-Treated or L-NAME-Treated Rats
The Phase 2 responses elicited by inj1-6 of FAD2.5 after injections of saline or L-NAME are summarized in Figure 3. Inj1 of FAD2.5 elicited pronounced decreases in MAP, MR, and HLRI and smaller decreases in HLRD in saline-treated rats. Subsequent injections of FAD2.5 elicited similar responses except that the decreases in HLRD gradually diminished upon repeated injection of FAD2.5. Inj1 of FAD2.5 in L-NAME-treated rats elicited similar responses to those of Inj1 in saline-treated rats. Inj2-6 of FAD2.5 elicited progressively smaller decreases in MAP, HLRI, and HLRD. The loss of vasodilation in the denervated hindlimb bed occurred more rapidly than in the sympathetically intact hindlimb bed.
Hemodynamic Responses Elicited by FAD0.25
The responses elicited by FAD0.25 in saline-treated rats were similar before and after inj1-6 of saline or FAD2.5 (P > 0.05, for all comparisons; data not shown). The Phase 1 responses elicited by FAD0.25 before and after inj1-6 of saline or FAD2.5 in L-NAME-treated rats are summarized in Figure 4. The FAD0.25 responses were similar before and after inj1-6 of saline. In contrast, the Phase 1 and 2 depressor and vasodilator responses elicited by FAD0.25 were markedly attenuated in rats, which received inj1-6 of FAD2.5. The Phase 1 increases in MR elicited by FAD0.25 were augmented in rats that received inj1-6 of FAD2.5 but not inj1-6 of saline.
Hemodynamic Responses Elicited by Acetylcholine and Bradykinin
The depressor and vasodilator responses elicited by acetylcholine before and after inj1-6 of saline or FAD2.5 in saline-treated or L-NAME-treated rats are summarized in Figure 5. Before injection of saline or L-NAME, acetylcholine elicited substantial decreases in MAP, MR, and HLRI and relatively minor decreases in HLRD. These responses were similar before and after inj1-6 of saline or FAD2.5 in saline-treated rats. The acetylcholine-induced responses were augmented in L-NAME-treated rats, which had received inj1-6 of saline. The augmentation was greater in sympathetically-denervated than in intact hindlimbs. In contrast, the acetylcholine-induced decreases in MAP, MR, and HLRI were substantially diminished in L-NAME-treated rats that received inj1-6 of FAD2.5. The vasodilation in the denervated bed of L-NAME-treated rats that received inj1-6 of FAD2.5 was markedly less than in the rats that received inj1-6 of saline.
The depressor and vasodilator responses elicited by bradykinin before and after inj1-6 of saline or FAD2.5 in saline or L-NAME-treated rats are summarized in Figure 6. Before injection of saline or L-NAME, bradykinin elicited a depressor response that was associated with a pronounced decrease in MR but relatively minor decreases in HLRI and HLRD (see Pre values). The depressor and vasodilator responses elicited by bradykinin in saline-treated rats were similar before and after inj1-6 of saline or FAD2.5. The bradykinin-induced responses were markedly augmented in L-NAME-treated rats, which received inj1-6 of saline. In contrast, the bradykinin-induced reductions in MAP, MR, and HLRI were markedly diminished in L-NAME-treated rats, which received inj1-6 of FAD2.5. Although the decrease in HLRD in the L-NAME-treated rats that received inj1-6 of FAD2.5 was equivalent to the pre-FAD2.5 response, this decrease in HLRD was markedly less than in L-NAME-treated rats that received inj1-6 of saline.
Hemodynamic Responses Elicited by L-S-Nitrosocysteine and Sodium Nitroprusside
The depressor and vasodilator responses elicited by L-SNC are summarized in Figure 7. Before injection of saline or L-NAME, L-SNC elicited prompt reductions in MAP, MR, HLRI, and HLRD (see Pre values). The depressor and vasodilator responses elicited by L-SNC in saline-treated rats were similar before and after inj1-6 of saline or FAD2.5. The L-SNC-induced responses were augmented in L-NAME-treated rats, which received inj1-6 of saline and equally augmented in L-NAME-treated rats, which received inj1-6 of FAD2.5.
The depressor and vasodilator responses elicited by SNP are summarized in Figure 8. Before administration of saline or L-NAME, SNP elicited significant decreases in MAP, MR, and HLRD but not HLRI (see Pre values). The depressor and vasodilator responses elicited by SNP in saline-treated rats were similar before and after inj1-6 of saline or FAD2.5. The SNP-induced responses were augmented in L-NAME-treated rats, which received inj1-6 of saline, and equally augmented in L-NAME-treated rats, which received inj1-6 of FAD2.5. The injection of SNP elicited a pronounced fall in HLRI after administration of L-NAME in rats, which received inj1-6 of saline or FAD2.5.
Changes in Baseline Parameters
Resting pre-injection MAP, MR, HLRI, and HLRD values in the group of rats that received inj1-6 of saline after an injection of saline (Group 1, see Table 1) were 104 ± 2 mm Hg, 35 ± 4 mm Hg/kHz, 46 ± 6 mm Hg/kHz, and 93 ± 8 mm Hg/kHz, respectively. HLRD was substantially higher than HLRI (+47 ± 8 mm Hg/kHz, P < 0.05) due to the possible downregulation of endothelial NOS in the denervated hindlimb.31 Resting values in the other treatment groups were similar to those described above (P > 0.05, for all comparisons). The percent changes in resting parameters that resulted from administration of 6 injections of saline or FAD2.5 in saline-treated or L-NAME-treated rats are summarized in Figure 9. Columns 1 and 2 show the percent changes in resting parameters elicited by the injection of saline before the 6 injections of saline or FAD2.5 (Pre-Inj1 columns). Columns 3 and 4 show the percent changes that resulted from administration of 5 injections of saline or FAD2.5 in these saline-treated rats (Pre-Inj6 columns). Columns 5 and 6 show the percent changes in resting parameters elicited by injection of L-NAME before the 6 injections of saline or FAD2.5 (Pre-Inj1 columns). Columns 7 and 8 represent the percent changes that resulted from the administration of 5 injections of saline or FAD2.5 in these L-NAME-treated rats (Pre-Inj6 columns). The injection of saline did not affect resting parameters in the rats that subsequently received inj1-6 of saline or FAD2.5 (see Pre-Inj1 columns). Resting hemodynamic values were not affected by inj1-6 of saline or inj1-6 of FAD2.5 (see Pre-Inj6 columns). The administration of L-NAME increased resting parameters in the rats that subsequently received inj1-6 of saline or FAD2.5 (see Pre-Inj1 columns). The increase in HLRI was greater than the increase in HLRD (P < 0.05, for both comparisons). Resting hemodynamic values were not further affected by administration of inj1-6 of saline (see Pre-Inj6 columns). In contrast, resting parameters gradually rose during inj1-6 of FAD2.5 in L-NAME-treated rats such that the values were substantially higher than those of L-NAME-treated rats that had received inj1-6 of saline.
Effects of Indomethacin
The changes in resting hemodynamic values elicited by co-administration of indomethacin and L-NAME were similar to those elicited by L-NAME alone (P > 0.05, for all comparisons, data not shown). In addition, indomethacin did not affect the loss of hemodynamic responses elicited by successive injections of FAD2.5 in L-NAME-treated rats. For example, the Phase 2 decreases in MR elicited by injections 1-6 of FAD were, −49 ± 6%, −44 ± 7%, −41 ± 8%, −36 ± 7%, −28 ± 4%, and −19 ± 4%, respectively. The responses elicited by inj4-6 were smaller than those elicited by inj1 (P < 0.05, for all comparisons).
An example of specific binding of 3H-4DAMP in endothelial membranes from a thoracic aorta of a rat that received L-NAME and 6 injections of FAD2.5 is shown in the top panel of Figure 10, and an example of a Scatchard plot is shown in the bottom panel of the figure. Binding was saturable, and nonspecific binding was less than 10% of total binding at all concentrations of 3H-4DAMP. A summary of binding characteristics of 3H-4DAMP in endothelial cell membranes from saline-treated or L-NAME-treated rats that received 6 injections of saline or FAD2.5 are summarized in Figure 11. As can be seen, the maximal densities of the high affinity and low affinity 3H-4DAMP binding sites (BmaxH and BmaxL, respectively) were substantially increased in endothelial cell membranes from L-NAME-treated rats. These values from L-NAME-treated rats that received 6 injections of FAD2.5 were similar to those L-NAME-treated rats that received 6 injections of saline. The dissociation constants of the high and low affinity sites (KDH and KDL, respectively) were not altered by L-NAME.
Hemodynamic Actions of FAD
The systemic injections of the 0.25 and 2.5 μmol/kg doses of FAD elicited depressor responses in conscious rats that were associated with vasodilator responses in the hindlimb vasculature and an initial vasoconstriction followed by a vasodilation in the mesenteric circulation. The responses elicited by the 2.5 μmol/kg dose of FAD were substantially greater than those elicited by the 0.25 μmol/kg dose. Ralevic et al6 provided evidence that the vasodilator actions of FAD in the isolated perfused rat mesenteric bed were dependent on an intact endothelium. Accordingly, the present finding that FAD elicited vasodilator responses in the sympathetically denervated hindlimb bed suggest that FAD may elicit endothelium-dependent responses in vivo. However, the responses elicited by FAD were smaller in the denervated than the sympathetically intact hindlimb bed; as reported previously,23,47 the vasodilator actions of acetylcholine and bradykinin were diminished in the denervated bed, whereas the vasodilator actions of L-SNC and SNP were not. As such, it is possible that a significant portion of the vasodilator actions of FAD involves inhibition of norepinephrine release from post-ganglionic sympathetic nerve terminals. However, other mechanisms may also contribute to diminished FAD-induced vasodilation in sympathetically denervated beds, including (1) downregulation of NOS and subsequent diminished synthesis/storage of S-nitrosothiols in the endothelium,23,26 (2) downregulation of FAD recognition sites on vascular endothelial cells, and (3) the loss of the active sympathetic NOS-positive vasodilator pathway in the lumbar chain.26,27 Moreover, the gradual loss of response to FAD in the denervated hindlimb bed may be due to the diminished capacity to store and/or regenerate preformed pools of nitrosyl factors. The downregulation of NOS and diminished release of endothelium-derived nitrosyl factors may also contribute to the elevated resistance in the denervated hindlimb bed.23,47
Ralevic et al6 also reported that the vasodilator actions of FAD in the isolated perfused rat mesenteric bed were due to activation of endothelial P2Y-purinoceptors and found that FAD did not elicit vasoconstrictor actions at any dose in this preparation. However, we demonstrated that the FAD-induced vasoconstriction in the mesenteric bed and vasodilation in the mesenteric and hindquarter beds of anesthetized rats were not affected by high doses of the P2Y/P2X-purinoreceptor antagonist, suramin.8 Interestingly, Romanenko et al48 reported that FAD elicited relaxant and constrictor actions in guinea pig taenia coli longitudinal smooth muscle. The finding that FAD elicits pronounced hemodynamic responses whereas FMN elicits negligible responses8 raises the possibility that the active moiety in FAD is adenine. We are currently trying to establish whether FAD activates the recently defined G protein-coupled adenine receptors9-11 and whether these receptors are responsible for the vasoconstrictor and vasodilator actions of FAD, although the lack of selective antagonists for these receptors is a major obstacle to these studies at present.
Loss of Response to FAD in L-NAME-Treated Rats
Resting hemodynamic parameters in the rats that received a bolus injection of saline followed by 6 bolus injections of saline or FAD remained constant throughout the experiments. Moreover, resting hemodynamic parameters in L-NAME-treated rats remained constant during the administration of 6 injections of saline. However, baseline MAP and vascular resistances rose gradually and quite substantially during the injections of FAD2.5 in L-NAME-treated rats. The increases in resting vascular resistances would be consistent with the progressive loss of release of endothelium-derived nitrosyl factors (see below). The vasodilator responses elicited by inj2-6 of FAD2.5 in the mesenteric and sympathetically-intact hindlimb beds of saline-treated rats were similar to those elicited by inj1. However, the vasodilator responses elicited by inj2-6 of FAD2.5 in the sympathetically denervated beds gradually diminished such that inj5 and inj6 caused slightly smaller responses. The gradual loss of response to FAD may involve a progressive downregulation of FAD recognition sites and/or diminished release of endothelial nitrosyl factors. Specifically, the stores of nitrosyl factors may be more readily depleted in the sympathetically denervated bed. As in anesthetized rats,8 the vasodilator responses elicited by FAD2.5 in conscious rats were not diminished after administration of L-NAME. This suggests that the vasodilator actions of FAD do not depend upon the release of newly synthesized nitrosyl factors. However, the depressor and vasodilator responses elicited by the FAD2.5 markedly diminished upon repeated injection in L-NAME-treated rats. The lack of effect of indomethacin on the use-dependent loss of vasodilation to FAD2.5 in L-NAME-treated rats suggests that increased production of endothelial vasoconstrictor prostanoids49 is not responsible for the diminished vasodilation.
It is possible that the gradual loss of response to FAD2.5 in L-NAME-treated rats was due to the increased ability of L-NAME to block NO synthesis over time the FAD2.5 injections were given. However, we have determined that the first injection of FAD2.5, given 40 min after administration of L-NAME (50 μmol/kg, n = 8 rats), which corresponds to the time at which the sixth injection of FAD2.5 was given in the present study, elicited robust hemodynamic responses that were equivalent to those elicited by the first injection of FAD2.5 in the present study (Lewis and Bates, unpublished observations). For example, the Phase 2 reduction in MAP elicited by the first injection of FAD2.5 given 40 min after the administration of L-NAME was −52 ± 5%, which was virtually identical to the reduction in MAP elicited by the first injection of FAD2.5 (−49 ± 3%) given 15 min after the administration of L-NAME in the present study. Moreover, subsequent injections of FAD2.5 also elicited progressively smaller depressor and vasodilator responses as observed in the present study. Taken together, it appears possible that the progressive reduction in the vasodilator responses to FAD2.5 observed in the present study involves the gradual depletion of preformed pools of nitrosyl factors rather than a temporal improvement in the ability of L-NAME to block NO synthesis.
The loss of response to FAD in L-NAME-treated rats may involve the desensitization and/or downregulation of FAD recognition sites on vascular endothelial cells. However, the finding that the vasodilator actions of the endothelium-dependent agonists, acetylcholine and bradykinin, were markedly diminished in L-NAME/FAD-treated rats, supports the possibility that the injections of FAD elicited the “use-dependent” depletion of preformed stores of endothelial nitrosyl factors that could not be replenished in the absence of NO synthesis. This possibility is supported by (1) our finding that the density of muscarinic M3 receptors was augmented after injection of L-NAME, which suggests that endogenous nitrosyl factors regulate the disposition of M3 receptors in the plasma membranes of endothelial cells, and (2) in vivo23 and in vitro24,25 evidence that endothelial cells contain preformed pools of nitrosyl factors with pharmacological properties similar to L-SNC.14,15 The finding that the vasodilator actions of L-SNC and SNP were exaggerated in these rats argues against the possibility that the loss of response to FAD was due to the downregulation of the mechanisms by which the EDRFs/EDHFs released by FAD exert their vasodilator responses.
Vascular endothelial cells in smaller arteries contain large numbers of cytoplasmic vesicles.50-56 Moreover, large numbers of these vesicles contain NOS in their cytosolic membranes.54-56 The presence of NOS would provide a mechanism by which S-nitrosothiols could be generated and ultimately stored in vesicles. A variety of cell types including blood vessels contain preformed pools of S-nitrosothiols and dinitrosyl iron complexes,18-22 and we have obtained direct evidence that S-nitrosothiols are stored in cytoplasmic vesicles within endothelial cells of rat and canine mesenteric and femoral arteries (unpublished observations). It would seem reasonable to suggest that these cytoplasmic vesicles are mobilized to exocytosis by increases in intracellular Ca2+ elicited by stimuli such as sheer stress and the activation of Go,q protein-coupled receptors such as FAD recognition sites, and muscarinic and bradykinin receptors. The finding that bradykinin causes NOS to translocate to the plasma membranes of endothelial cells57 is consistent with a Ca2+-dependent mobilization and fusion of NOS-positive vesicles to the plasma membranes. Evidence that the plasma membranes of endothelial cells contain fusion proteins that promote vesicular exocytosis58 supports the possibility of exocytotic mechanisms in endothelial cells. The finding that calmodulin antagonists inhibit endothelium-dependent relaxations in arteries treated with NOS inhibitors by inhibiting the release rather than the mechanisms of action of EDHFs and/or EDRFs59,60 suggests that, similar to other secretory cells,21,61 Ca2+-calmodulin complex may initiate the mobilization of S-nitrosothiol-containing vesicles in endothelial cells. Moreover, the findings that inhibitors of mitochondrial electron transport, F1-ATPase or oxidative phosphorylation inhibit endothelium-dependent relaxations via the release of EDRF62-64 further suggests that exocytosis of vesicular stores of nitrosyl factors occurs in endothelial cells because these inhibitors block vesicular exocytosis in nerve terminals and adrenal chromaffin cells.65 Taken together, the above findings suggest that FAD may mobilize vesicles that contain preformed nitrosyl factors and that repeated application of FAD in L-NAME-treated rats induces a depletion of these vesicular nitrosyl factors because the blockade of NO synthesis will preclude the regeneration of these factors.
The responses elicited by acetylcholine, bradykinin, L-SNC, and SNP in saline and in L-NAME-treated conscious rats were similar to previous reports.23,26,47 NO-donors such as SNP elicit minimal responses in the intact hindlimb beds of conscious naïve rats, whereas acetylcholine and L-SNC elicit robust responses.23,26,47 This and other evidence23,26,47 supports in vitro and in vivo evidence that L-SNC is an EDRF/EDHF12-15 that elicits its vasodilator actions by mechanisms distinct from those of NO.14,15 The finding that the vasodilator actions of acetylcholine and bradykinin were not diminished in rats treated with L-NAME only, is consistent with previous findings that multiple injections of these endothelium-dependent agonists were needed to elicit a loss of response.23 Moreover, the present and published findings23 that the vasodilator actions of bradykinin were augmented in rats treated with L-NAME only is consistent with evidence that S-nitrosothiols markedly reduce the coupling of bradykinin receptors to Gi and Gq proteins.66
Vesicular exocytosis is a vital function of endothelial cells, and factors such as haemostatic proteins such as von Willebrand Factor,67 ATP,68 immunoglobulin G,69 ANP,70 endothelin,71 and histamine72 undergo exocytosis. The present study provides evidence that systemic injections of FAD elicit pronounced vasodilator responses in conscious rats, which may be due to the release of preformed stores of nitrosyl factors such as L-SNC from endothelial cells. The mechanisms by which FAD elicits vasoconstriction remain to be determined but do involve vasoconstrictor prostanoids (present study), α1-adrenoceptors, angiotensin AT1 receptors, arginine vasopressin receptors, or 5-HT2 receptors (unpublished findings). The present findings also support substantial evidence that, whereas L-NAME inhibits muscarinic receptor binding in some isolated cell systems, it does not diminish muscarinic receptor signaling in vitro14,15 or in vivo.23 Rather, our findings raise the possibility that endogenous nitrosyl factors regulate the disposition (density) of muscarinic M3 receptors in vascular membranes.
1. Flashner MS, Massey V. Flavoprotein oxygenases. In: Hayaishi O, ed. Molecular Mechanisms of Oxygen Activation
., New York, NY: Academic Press; 1974:245-260.
2. Mathews CK, Van Holde KE. Biochemistry
. California: The Benjamin/Cummings Publishing Company; 1990:474-478.
3. Hustad S, McKinley MC, McNulty H, et al. Riboflavin, flavin mononucleotide, and flavin adenine dinucleotide
in human plasma and erythrocytes at baseline and after low-dose riboflavin supplementation. Clin Chem
4. Cicek G, Schiltz E, Staiger J, et al. Specific stimulation of peripheral blood mononuclear cells from patients with acute myocarditis by peptide-bound flavin adenine dinucleotide
(FAD), a naturally occurring autologous hapten. Clin Exp Immunol
5. Kotegawa M, Sugiyama M, Haramaki N. Protective effects of riboflavin and its derivatives against ischemic reperfused damage of rat heart. Biochem Mol Biol Int
6. Ralevic V, Hoyle CH, Burnstock G. Pivotal role of phosphate chain length in vasoconstrictor versus vasodilator actions of adenine dinucleotides in rat mesenteric arteries. J Physiol
7. Moncada S. Adventures in vascular biology: a tale of two mediators. Philos Trans R Soc Lond B Biol Sci
8. Hashmi-Hill MP, Graves JE, Sandock K, et al. Hemodynamic responses elicited by systemic injections of flavin adenine dinucleotide
in anesthetized rats
. J Cardiovasc Pharmacol
9. Bender E, Buist A, Jurzak M, et al. Characterization of an orphan G protein-coupled receptor localized in the dorsal root ganglia reveals adenine as a signaling molecule. Proc Natl Acad Sci USA
10. Gorzalka S, Vittori S, Volpini R, et al. Evidence for the functional expression and pharmacological characterization of adenine receptors in native cells and tissues. Mol Pharmacol
11. Watanabe S, Ikekita M, Nakata H. Identification of specific [3
H]adenine-binding sites in rat brain membranes. J Biochem
12. Myers PR, Minor RL, Guerra R, et al. Vasorelaxant properties of the endothelium-derived relaxing factor more closely resemble S-nitrosocysteine than nitric oxide. Nature
13. Rosenblum WI. Endothelium-derived relaxing factor in brain blood vessels is not nitric oxide. Stroke
14. Batenburg WW, Popp R, Fleming I, et al. Bradykinin-induced relaxation of coronary microarteries: S-nitrosothiols as EDHF? Brit J Pharmacol
15. Batenburg WW, De Vries R, Saxena PR, et al. L-S-nitrosothiols: endothelium-derived hyperpolarizing factors in porcine coronary arteries? J Hypertens
16. Buyukafsar, K, Gocmen C, Secilmis A, et al. Evidence that the nitrergic neurotransmitter and endothelium-derived relaxing factor might be S-nitrosothiols in the mouse corpus cavernosum. Acta Med Okayama
17. Ignarro LJ. Nitric oxide: A novel signal transduction mechanism for transcellular communication. Hypertension
18. Muller B, Kleschyov AL, Alencar JL, et al. Nitric oxide transport and storage in the cardiovascular system. Ann NY Acad Sci
19. Vanin AF. Dinitrosyl iron complexes and S-nitrosothiols are two possible forms for stabilization and transport of nitric oxide in biological systems. Biochemistry
20. Kluge I, Gutteck-Amsler U, Zollinger M, et al. S-nitrosoglutathione in rat cerebellum: identification and quantification by liquid chromatography-mass spectrometry. J Neurochem
21. Chvanov M, Gerasimenko OV, Petersen OH, et al. Calcium-dependent release of NO from intracellular S-nitrosothiols. EMBO J
22. Jia L, Bonaventura C, Bonaventura J, et al. S-nitrosohemoglobin: a dynamic activity of blood involved in vascular control. Nature
23. Davisson RL, Bates JN, Johnson AK, et al. Use-dependent loss of acetylcholine- and bradykinin-mediated after nitric oxide synthase inhibition: Evidence for preformed stores of nitric oxide-containing factors in vascular endothelial cells. Hypertension
24. Danser AHJ, De Vries R, Schoemaker RG, et al. Bradykinin-induced release of nitric oxide by the isolated perfused rat heart: importance of preformed pools of nitric-oxide containing factors. J Hypertens
25. Danser AJ, Tom B, De Vries R, et al. L-NAME-resistant bradykinin relaxation in porcine coronary arteries is NO-dependent. Effect of ACE inhibition. Brit J Pharmacol
26. Davisson RL, Shaffer RA, Johnson AK, et al. Stimulation of lumbar sympathetic nerves may produce hindlimb vasodilation via the release of pre-formed stores of nitrosyl factors
27. Davisson RL, Possas OS, Murphy SP, et al. Neurogenically-derived nitrosyl factors
mediate lumbar sympathetic vasodilation in the hindlimb vasculature of the rat. Am J Physiol
28. Venturini CM, Palmer RMJ, Moncada S. Vascular smooth muscle contains a depletable store of a vasodilator which is light-activated and restored by donors of nitric oxide. J Pharmacol Exp Ther
29. Buyukafsar K, Demrel-Yilmaz E, Gocmen C, et al. Depletion and restoration of the putative photosensitive stores of nitric oxide in the isolated mouse gastric fundus. J Pharmacol Exp Ther
30. Buyukafsar K, Levent A, Un I, et al. Mediation of nitric oxide from photosensitive stores in the photorelaxation of the rabbit corpus cavernosum. Eur J Pharmacol
31. McCormack DG, Mak JC, Minette P, et al. Muscarinic receptor subtypes mediating vasodilation in the pulmonary artery. Eur J Pharmacol
32. Dauphin F, Hamel E. Muscarinic receptor subtype mediating vasodilation feline middle cerebral artery exhibits M3 pharmacology. Eur J Pharmacol
33. Altiere RJ, Travis DC, Roberts J, et al. Pharmacological characterization of muscarinic receptors mediating acetylcholine-induced contraction and relaxation in rabbit intrapulmonary arteries. J Pharmacol Exp Ther
34. Hammarstrom AK, Parkington HC, Coleman HA. Release of endothelium-derived hyperpolarizing factor (EDHF) by M3 receptor stimulation in guinea-pig coronary artery. Brit J Pharmacol
35. Eglen RM, Reddy H, Watson N. Selective inactivation of muscarinic receptor subtypes. Int J Biochem
36. Cheng L, Yang C, Hsu L, et al. Acute exercise enhances receptor-mediated endothelium-dependent vasodilation by receptor up-regulation. J Biomed Sci
37. Lewis SJ, Hashmi-Hill MP, Owen JR, et al. The vasodilator potency of the endothelium-derived relaxing factor, L-S-nitrosocysteine, is impaired in Spontaneously Hypertensive rats
. Vasc Pharmacol
38. Lewis SJ, Hashmi-Hill MP, Owen JR, et al. ACE inhibition restores the vasodilator potency of the endothelium-derived relaxing factor, L-S-nitrosocysteine, in Spontaneously Hypertensive rats
. Vasc Pharmacol
39. Whalen EJ, Johnson AK, Lewis SJ. Functional evidence for the rapid desensitization of 5-HT3
receptors on vagal afferents mediating the Bezold-Jarisch reflex. Brain Res
40. Lacolley P, Owen JR, Bates JN, et al. Tachyphylaxis to 5-HT3
receptor-mediated activation of vagal afferents is prevented by co-activation of 5-HT2
receptors. Brain Res
41. Haywood JR, Shaffer RA, Fastenow C, et al. Regional blood flow measurement with pulsed Doppler flowmeter in conscious rat. Am J Physiol
42. Lowry OH, Rosebrough NJ, Farr AL, et al. Protein measurement with the Folin phenol reagent. J Biol Chem
43. Michel AD, Stefanich E, Whiting RL. Direct labeling of rat M3-muscarinic receptors by [3
H]4DAMP. Eur J Pharmacol
44. Lewis SJ, Verberne AJM, Summers RJ, et al. Reduced glutamate binding in rat dorsal vagal complex after nodose ganglionectomy. Brain Res Bull
45. Winer BJ. Statistical Principles of Experimental Design
. New York, NY: McGraw Hill Company; 1971:752-809.
46. Wallenstein S, Zucker CL, Fleiss JL. Some statistical methods useful in circulation research. Circ Res
47. Travis MD, Davisson RL, Bates JN, et al. Hemodynamic effects of L- and D-S-nitroso-β,β-dimethylcysteine in the rat. Am J Physiol
. 1997;273: H1493-H1501.
48. Romanenko AV, Khalmuradov AG, Shuba MF. Mechanism of action of vitamin B1, thiamine pyrophosphate, and FAD on smooth muscle cells. Ukr Biokhim Zh
49. Noll G, Lang MG, Tschudi MR, et al. Endothelial vasoconstrictor prostanoids modulate contractions to acetylcholine and ANG II in Ren-2 rats
. Am J Physiol
50. Wagner RC, Casley-Smith JR. Endothelial vesicles. Microvasc Res
51. Huttner I, Gabbiani G. Vascular endothelium in hypertension. In: Genest J, Kuche O, Hamet P, Cantin M, eds. Hypertension
. New York, NY: McGraw-Hill; 1983:473-88.
52. Lane NJ, Revest PA, Whytock S, et al. Fine-structural investigation of rat brain microvascular endothelial cells: tight junctions and vesicular structures in freshly isolated and cultured preparations. J Neurocyt
53. Moldovan NI, Heltianu C, Simionescu N, et al. Ultrastructural evidence of differential solubility in Triton X-100 of endothelial vesicles and plasma membrane. Exp Cell Res
54. Loesch A, Belai A, Burnstock G. Ultrastructural localization of NADPH-diaphorase and colocalization with nitric oxide synthase in endothelial cells of the rabbit aorta. Cell Tissue Res
55. Loesch A, Belai A, Burnstock G. An ultrastructural study of NADPH-diaphorase and nitric oxide synthase in perivascular nerves and vascular endothelium of rat basilar artery. J Neurocyt
56. Loesch A, Burnstock G. Ultrastructural study of perivascular nerve fibres and endothelial cells of the rat basilar artery immunolabelled with monoclonal anti-bodies to neuronal and endothelial nitric oxide synthase. J Neurocytol
57. Michel T, Li GK, Busconi L. Phosphorylation and subcellular translocation of endothelial nitric oxide synthase. Proc Natl Acad Sci USA
58. Schnitzer JE, Liu J, Oh P. Endothelial caveolae have the molecular transport machinary for vesicle budding, docking, and fusion including VAMP, NSF, SNAP, Annexins, and GTPases. J Biol Chem
59. Illiano SC, Nagao T, Vanhoutte PM. Effect of calmodulin inhibitors on endothelium-dependent relaxations to bradykinin in porcine and canine coronary arteries. In: Moncada, S, Marletta MA, Hibbs JB, eds. The Biology of Nitric Oxide
. London: Portland Press; 1993:143-52.
60. Nagao T, Illiano S, Vanhoutte PM. Calmodulin antagonists inhibit endothelium-dependent hyperpolarization in the canine coronary artery. Brit J Pharmacol
61. Thomas-Reetz, AC, De Camilli P. A role for synaptic vesicles in non-neuronal cells: clues from pancreatic beta cells and from chromaffin cells. FASEB J
62. Griffith TM, Edwards DH, Lewis MJ, et al. Production of endothelium-derived relaxing factor is dependent on oxidative phosphorylation and extracellular calcium. Cardiovasc Res
63. Richards JM, Gibson IF, Martin W. Effects of hypoxia and metabolic inhibitors on production of prostacyclin and endothelium-derived relaxing factor by pig endothelial cells. Brit J Pharmacol
64. Weir CJ, Gibson IF, Martin W. Effects of metabolic inhibitors on endothelium-dependent and endothelium-independent dilation of rat and rabbit aorta. Brit J Pharmacol
65. Thureson-Klein AK, Klein RL. Exocytosis from neuronal large dense-cored vesicles. Int Rev Cytol
66. Miyamoto A, Laufs U, Pardo C, et al. Modulation of bradykinin receptor ligand binding affinity and its coupled G-proteins by nitric oxide. J Biol Chem
67. Van Mourik JA, Romani De Wit T, Voorberg J. Biogenesis and exocytosis of Weibel-Palade bodies. Histochem Cell Biol
68. Bodin P, Burnstock G. Evidence that release of adenosine triphosphate from endothelial cells during increased shear stress is vesicular. J Cardiovasc Pharmacol
69. Ober RJ, Martinez C, Lai X, et al. Exocytosis of IgG as mediated by the receptor, FcRn: An analysis at the single-molecule level. Proc Natl Acad Sci USA
70. Newman TM, Severs NJ, Skepper JN. The pathway of atrial natriuretic peptide release from cell to plasma. Cardioscience
71. Harrison VJ, Barnes K, Turner AJ, et al. Identification of endothelin 1 and big endothelin 1 in secretory vesicles isolated from bovine aortic endothelial cells. Proc Natl Acad Sci USA
72. Ozaka T, Katsuki M, Fukushige H, et al. Histamine release from Weibel-Palade bodies of toad aortas induced by endothelin-1 and sarafotoxin-S6b. Anat Rec