| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
LOCAL CONTROL OF CIRCULATION
1Department of Physiology and Pharmacology, West Virginia University School of Medicine, Morgantown, West Virginia 26506-9229; and 2Department of Physiology and Biophysics, University of Mississippi Medical Center, Jackson, Mississippi 39216-4505
Submitted 2 August 2002 ; accepted in final form 13 March 2003
 |
ABSTRACT |
|---|
 TOP ABSTRACT MATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. ATP (100 µM) resulted in
reproducible increases in bath concentration from 73 ± 22 to 279
± 50 pg/ml (P < 0.05, n = 5). This response was
abolished by indomethacin (10 µM, P < 0.05). To ascertain
whether the endothelium was the source of prostacyclin, endothelium was
disrupted using air (n = 10) or deoxycholic acid (n = 6).
Perfusion with air significantly reduced (P < 0.05) but did not
completely abolish ATP-stimulated release of prostacyclin, while deoxycholic
acid totally abolished the response (P < 0.05). The nonselective
P2 receptor antagonist reactive blue 2 (100 µM) attenuated ATP-mediated
release of prostacyclin but did not significantly alter ACh-stimulated release
of prostacyclin. The nonselective adenosine receptor antagonist xanthine amine
congener (1 µM) had no effect on ATP-stimulated release, and adenosine did
not stimulate the release of prostacyclin. These results show that increases
in intraluminal concentration of ATP stimulate abluminal release of
prostacyclin from the venous endothelium. This effect is mediated by P2
receptors while adenosine and its receptors are not involved in this
response. prostaglandins; P2Y receptor; cyclooxygenase
Once released from cells, extracellular ATP is rapidly hydrolyzed to ADP, AMP, and adenosine via extracellular ectonucleotidases. The fate of extracellular ATP has been reviewed elsewhere (3, 13, 28). ATP and ADP mediate their vasodilatory effects via P2Y receptors while adenosine and AMP mediate their vasodilatory effects via adenosine (P1) receptors. Subtypes of both receptors have been identified on vascular endothelial cells (22). The relative contribution of P2Y and P1 receptors to ATP-mediated vasodilation remains controversial, and thus a second aim of the present study was to determine whether ATP-stimulated prostacyclin release was mediated via activation of P2Y or P1 receptors.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
General Procedures
Male golden hamsters (140–200 g) were anesthetized with pentobarbital sodium (60 mg/kg ip). The hindlimb vasculature was exposed by an incision of the skin, and the tissues were kept moist by superfusing them with warm physiological salt solution (PSS) of the following composition (in mM): 131.9
NaCl, 4.7 KCl, 2.0 CaCl2, 1.2 MgSO4, and 20
NaHCO3, equilibrated with 95% N2-5% CO2. With
the use of an operating microscope, the superficial epigastric vein was
separated from the adjacent superficial epigastric artery and surrounding
tissue, and small branches were ligated using 10–0 nylon suture.
Attempts were made to obtain as great a length of vein as possible; however,
branching of this vessel into three smaller vessels occurred within 5 mm of
its origin at the femoral vein, limiting the length of the vessels in situ to
3–5 mm. Using a catheter containing MOPS-buffered PSS (composition in
mM: 145 NaCl, 4.7 KCl, 2.0 CaCl2, 1.17 MgSO4, 1.2
NaH2PO4, 0.02 EDTA, 3.0 MOPS, 5.0 glucose), the vein was
cannulated at its proximal end, removed from the animal, and flushed of blood.
After transferring the vessel to a 1-ml bath containing warm (37°C)
MOPS-PSS, a second catheter was placed in the distal end of the vessel, and
perfusion with MOPS-PSS at 37°C, bubbled with CO2, was
commenced. Vessels were perfused from a reservoir at a fixed height and
constant pressure (20 mmHg static pressure, which resulted in a flow rate of
1 ml/min). Because the majority of the resistance of this system was in
the tubing before the vessel, any changes in vessel diameter, which may have
occurred during the experiments, were unlikely to have any significant effect
on perfusion pressure and thus were unlikely to be of any functional
significance in the present study.
At the completion of the experimental protocols (described below), the vessels were removed from the catheters and fixed with 10% formalin. Vessels were then paraffin embedded, cut into 5-µm sections, and stained with hematoxylin and eosin. Vessels were viewed under a light microscope to confirm the presence of endothelium.
Experimental Protocols
Values of n are the number of vessels used for a given protocol. Different protocols were undertaken when two vessels were harvested from the same animal. Only one protocol was carried out in each vessel. Vessels were equilibrated for 1 h with collection of bath solution commencing at 30 min.
The entire 1-ml bath content was collected and replaced with fresh MOPS-PSS
every 10 min for the duration of the experiment. Samples were stored at
-20°C until they were assayed, in triplicate, for the stable metabolite of
prostacyclin, 6-keto-PGF1, using the Elisa
Technologies 6-keto-PGF1 kit (Neogen, Lexington,
KY; product 404310). This technique has an assay range of 20–2,000
pg/ml, with an intra- and interassay coefficient of variation of 10%. An
increase in the amount of 6-keto-PGF1 measured in
the bath solution was interpreted as an increase in the amount of prostacyclin
released from the veins. In all of the following experiments, ATP (100 µM),
ACh (10 µM), or adenosine (100 µM) was perfused for 30 min to establish
a control response. After a washout period of 10 min, during which time the
vessels were perfused with regular MOPS-PSS, the veins were treated with an
inhibitor (which remained in the perfusate solution until the end of the
experiment) or the endothelium was disrupted using one of two methods
described below. Vessels were then exposed to the same agonist used to
establish the control response.
Protocol 1: cyclooxygenase inhibition. The effect of cyclooxygenase inhibition on ATP-mediated prostacyclin release was examined by perfusing the vessels with indomethacin (10 µM, n = 6) for 20 min before establishing a second response to ATP. Time and vehicle (sodium carbonate, 10 mM) controls were performed in additional vessels (n = 5 and n = 6, respectively).
Protocol 2: endothelial disruption. The endothelium was disrupted in one of two ways. First, an air bubble was perfused through the vessel for several seconds before establishing the second response to ATP (n = 10). The second protocol utilized the detergent deoxycholic acid. This was backflushed through the vein immediately after the first exposure to ATP (n = 6). Vessels were then allowed to stabilize for 20 min before perfusing with ATP for a second time.
Protocol 3: purine (P2) receptor blockade. To determine whether ATP was mediating its effect via P2 receptors, the nonselective antagonist reactive blue 2 (100 µM, n = 6) was added to the perfusate solution 20 min before the second exposure to ATP. To determine that the effect of this antagonist was specific for ATP, these experiments were repeated in different vessels using ACh in place of ATP (n = 5).
Protocol 4: adenosine receptor blockade. To determine whether the ATP-mediated release of prostacyclin was due to the hydrolysis of ATP to adenosine, experiments were performed whereby adenosine (100 µM) was used instead of ATP. The P1 receptor antagonist xanthine amine congener (XAC, 1
µM) was added to the perfusate before the second exposure to adenosine
(n = 5). XAC has been reported to block the A1 and
A2 subtypes of the P1 receptor
(10). In additional vessels
(n = 4), we examined the effect of XAC (1 µM) on bath
concentrations of 6-keto-PGF1, in response to
ATP, to determine whether other metabolites of ATP that act at P1 receptors
were responsible for prostacyclin release. As with all of the protocols
utilizing inhibitors, XAC was added to the perfusate 20 min before the second
exposure to ATP and remained in the solution for the duration of the
experiment.
Drugs
All drugs were purchased from Sigma (St. Louis, MO). ATP was dissolved in distilled water (100 mM), and XAC was dissolved in DMSO (100 µM). Both drugs were stored in aliquots at -20°C and thawed on the day of the experiment. Indomethacin was dissolved in sodium carbonate (10 mM) and stored in the refrigerator. Adenosine was dissolved and diluted in MOPS-PSS on the day of the experiment, and aliquots of ATP, indomethacin, and XAC were diluted in MOPS-PSS on the day of the experiment. Deoxycholic acid (3%) was suspended in distilled water. On the day of the experiment, the solution was heated to 37°C before back-flushing the vessel.
Analytic and Statistical Methods
As each vessel was exposed to ATP or adenosine before and after an experimental protocol, the effects of ATP and of the experimental protocol on bath concentrations of 6-keto-PGF1 were analyzed using a two-way repeated measures ANOVA. All data are means ± SE.
Statistical significance was accepted at P < 0.05.
|
RESULTS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. In the absence of ATP, basal levels
of 6-keto-PGF1 were detected in the bath solution
(172 ± 35 pg/ml, n = 7). ATP (1, 10, and 100 µM) resulted
in dose-dependent increases in the amount of
6-keto-PGF1 measured (213 ± 37, 322
± 63, 680 ± 216 pg/ml, respectively, n = 7); however,
due to the large variability in basal levels of release, statistical
significance was only achieved at 100 µM. Thus this concentration was used
throughout this study. Figure
2A demonstrates that ATP (100 µM)-stimulated release
of prostacyclin was reproducible (n = 5); however, it is interesting
to note that the second challenge to ATP was met with a decreased response if
a minimum of 30 min was not allowed between ATP challenges (data not shown).
The ATP-stimulated increase in bath concentration of
6-keto-PGF1 was abolished by intraluminal
administration of indomethacin (Fig.
2B, P < 0.05, n = 6). The vehicle
for indomethacin, sodium carbonate, had a small but significant effect on
ATP-stimulated release of prostacyclin
(Fig. 2C, P
< 0.05, n = 6).
|
|
The effect of endothelial disruption on 6-keto-PGF1 release is shown in Fig. 3. ATP (100 µM) was associated with an increase in 6-keto-PGF1
concentration from 160 ± 32 to 653 ± 139 pg/ml
(Fig. 3A, P
< 0.05, n = 10). Disruption of the endothelium using air bubbles
significantly decreased ATP-mediated prostacyclin release but did not
completely abolish this response (from 144 ± 33 to 421 ± 102
pg/ml, n = 10, Fig.
3A). However, when the venous endothelium was disrupted
using deoxycholic acid, the response to ATP was totally abolished
(Fig. 3B, P
< 0.05, n = 6).
|
The nonselective P2 receptor antagonist reactive blue 2 (100 µM) was shown to significantly decrease ATP-stimulated increases in 6-keto-PGF1 concentration (from 537 ± 111 to 205 ± 82 pg/ml, P < 0.05, n = 6,
Fig. 4A). In contrast,
100 µM reactive blue 2 was without effect on ACh-stimulated increases in
6-keto-PGF1 concentration
(Fig. 4B, n =
5).
|
ATP-stimulated increases in 6-keto-PGF1 were not altered by the adenosine (P1) receptor antagonist XAC (1 µM, n = 4, Fig. 5A), and adenosine (100 µM) was not associated with any increase in bath
concentration of 6-keto-PGF1 from these veins
(Fig. 5B, n =
5).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
,
indicating large ATP-mediated increases in prostacyclin release from these
vessels. The response to ATP was blocked by indomethacin and was partially or
totally blocked by damage to the venous endothelium. These results provide
direct evidence that an increase in intraluminal ATP concentration can
stimulate prostacyclin release from endothelial cells situated on the venous
side of the circulation. We also sought to clarify whether ATP was exerting its effect via direct stimulation of purine P2 receptors or whether a metabolite such as adenosine was mediating the effect. The results show that the nonselective P2 antagonist reactive blue 2 did inhibit ATP-mediated prostacyclin release. In addition, the adenosine receptor antagonist XAC had no effect on the response elicited by ATP while direct application of adenosine did not stimulate the release of prostacyclin. These results rule out a role for adenosine receptors in mediating this effect and suggest that ATP is exerting its effect via direct stimulation of P2 receptors on the venular endothelium.
Increases in skeletal muscle metabolism are associated with increases in the diameter of arterioles supplying the muscle. Studies from our laboratory have consistently demonstrated that electrical field stimulation of the hamster cremaster muscle results in dilation of the cremasteric arterioles. In larger arterioles paired with a venule, this response is dependent on an intact venular endothelium (23) and involves metabolites of arachidonic acid (17, 21). Most recently, we established that inhibition of cyclooxygenase, with indomethacin, only attenuated the functional dilation of arterioles that were paired with venules (14). These studies collectively suggest that increases in skeletal muscle metabolism are associated with the release of a cyclooxygenase product from the venular endothelium, which can influence the diameter of the adjacent arteriole.
In 1995, Ellsworth and colleagues (9) demonstrated that red blood cells could release ATP under hypoxic and low-pH conditions. As such conditions occur in the venous circulation during exercise, we hypothesized that ATP might play a role in functional hyperemia. We (15) have shown that ATP infused into a venule running parallel to an arteriole results in dilation of the cremasteric arteriole. This effect of ATP was shown to be dependent on an intact venular endothelium and was inhibited when indomethacin was applied globally to the entire cremaster preparation. These studies support a role for ATP in control of vascular tone and further support our hypothesis that cyclooxygenase metabolites from the venular endothelium can influence arteriolar diameter. Furthermore, a recent study demonstrated that venous plasma ATP concentration progressively increased with incremental exercise, suggesting that circulating ATP levels were tightly coupled to the oxygenation of hemoglobin in healthy humans (12).
In the present study, we aimed to provide direct evidence for ATP-stimulated release of prostacyclin from veins. Veins were cannulated at both ends and were mounted in a 1-ml chamber containing PSS. This setup allowed us to administer the ATP intraluminally and measure the prostacyclin release abluminally. Addition of ATP to the perfusion solution resulted in the release of prostacyclin into the bath solution. These results are consistent with our previous work (5) where we showed that ATP-mediated prostacyclin release from small veins was a calcium-dependent response. We further confirmed that we were measuring a metabolite of the cyclooxygenase pathway by blocking the response with indomethacin.
Red blood cells contain millimolar concentrations of ATP (18). Although recent studies have measured venous plasma concentrations of ATP in the range of 6–7 µM during exercise (12), it is questionable as to whether levels such as that used in our study would be achieved under normal physiological conditions. However, we have demonstrated in vivo (hamster cremaster) that venular infusions of 100 µM ATP (15) resulted in diameter changes comparable to those observed during contraction of the same muscle (17). The assay used in the present study produced highly variable results, and we chose a relatively high concentration of ATP in the hope that we would observe statistically significant results. This concentration of ATP consistently resulted in the release of prostacyclin from the veins.
It is not known whether the amount of prostacyclin released from these veins is comparable to that released in vivo from venules during muscle contraction. We have performed a simple analysis to estimate the physiological relevance of the prostacyclin release. In the current experiments the veins
were 0.5 mm in diameter and 5 mm in length. Assuming a cylindrical shape,
this would lead to a surface area of 8 mm2. The total
diffusion volume of these vessels is the 1,000 mm3 (1 ml) of the
bath. For in situ venules, we could perform the same calculation of a 1-mm
long, 0.1-mm in diameter venule with an effective diffusion distance 0.1 mm on
either side. This would lead to a surface area of 0.3 mm2 and a
total diffusion volume of 0.07 mm3. One can see that for the small
venules the ratio of volume to surface area is quite small, suggesting that
the effective prostacyclin concentration around venules may actually be higher
than in the isolated veins. Whether this response is physiologically relevant
is difficult to answer. Iloprost, a stable analog of prostacyclin, will cause
a vasodilation of arterioles at a concentration of 1,000 pg/ml
(11). The current studies show
6-keto-PGF1, increasing from 100
to 600 pg/ml. With similar vasoactive responses between iloprost and
prostacyclin (J. C. Frisbee, personal communication), the analysis suggests
that the prostacyclin concentration around the venules would be even higher,
which would cause a vasodilation. Until we are able to directly measure the
6-keto-PGF1 release from the venules, it will be
difficult to answer this question.
In our previous study, although it was highly probable, we did not confirm that the 6-keto-PGF1 measured in the bath solution was coming from the endothelium. Thus, in the present study, we attempted to remove the endothelium by briefly perfusing the vein with an air
bubble. It was surprising to us that this perturbation only reduced the
ATP-stimulated prostacyclin release by 50%. However, in a previous study
of functional dilation in the hamster cremaster muscle, we showed that
infusions of air into venules only reduced the functional response by 50%
(23). This raises the question
of whether air bubbles are effective tools for denuding or damaging venous
endothelium. In the present study, endothelial cell nuclei were identified in
the histological sections of veins after perfusion with air bubbles. Thus a
reduced response after an air bubble may reflect a reduced number of
endothelial cells rather than complete denudation. Unfortunately, the
histological techniques used in the present study did not allow for an
accurate quantification of the number of endothelial cells present; thus this
suggestion is entirely speculative. Alternatively, it is possible that the
mechanism via which air bubbles inactivate endothelial cells is not by removal
but rather by dehydration. Again, the partial attenuation of responses after
air bubbles may reflect only partial damage to the endothelium.
To further examine the contribution of the endothelium to the ATP-mediated release of prostacyclin, we used deoxycholic acid to inactivate the endothelial cells. Deoxycholic acid has been reported to remove membrane proteins without actually removing the endothelial cells
(26), thus rendering them
inactive. Studies using deoxycholic acid have demonstrated that this compound
is very effective at eliminating endothelium-dependent relaxation without
altering endothelium-independent relaxation
(1,
7,
8,
24). Furthermore,
Cusma-Pelogia and colleagues
(8) showed that inactivation of
the endothelium using this substance did not alter phenylephrine- or
norepinephrine-induced constriction nor did it alter the morphology of the
medial smooth muscle cells. Under light microscopy we confirmed that
endothelial cells were still present after treatment with deoxycholic acid and
that the smooth muscle layer was still intact. However, after the
administration of deoxycholic acid, ATP-mediated
6-keto-PGF1 release was completely abolished.
These studies confirm that ATP-mediated prostacyclin release from hamster
epigastric veins is endothelium-dependent and are consistent with studies
using isolated arterial preparations
(2) and cultured endothelial
cells (4,
5,
20).
Throughout these experiments, we were faced with large variations in basal levels of 6-keto-PGF1. This was particularly evident in the group of vessels used for the air perfusion. It is likely that among the other groups of vessels, variations in basal concentrations of
6-keto-PGF1 were due to differences in the number
of functional endothelial cells. It would be reasonable to expect that factors
such as vessel length, vessel diameter, trauma during vessel harvesting, and
health of the animals may have contributed to such variations. It is difficult
to attribute the very large differences observed in the air perfusion group of
vessels to these factors; however, an alternative explanation is not readily
available.
ATP is broken down rapidly in the circulation via the catalytic actions of ectonucleotidases in the endothelium. In the coronary circulation, ATP is broken down almost instantaneously with adenosine being one of the prominent metabolites (16). The relative contributions of ATP and adenosine to vasodilation are still controversial, and it is possible that the effect of ATP on the release of prostacyclin from veins is, in part, mediated via P1 (adenosine) receptors. To discriminate between P1 and P2 receptors, we utilized the nonselective P2 receptor antagonist reactive blue 2 (22). This antagonist effectively inhibited ATP-mediated prostacyclin release, suggesting that P2 receptors are involved in this response. Reactive blue 2 does not discriminate between P2X and P2Y receptors, and despite the large amount of evidence showing that vasodilatory endothelial P2 receptors fall into the P2Y category (22), P2X receptors may also be involved in endothelium-dependent vasodilation. P2X4 receptors have been identified on vascular endothelial cells, and activation of these receptors by ATP results in increases in intracellular calcium (27). While the results of our study do not discriminate between P2X and P2Y receptor subtypes, it is interesting to note that the refractory period, or desensitization effect after the first administration of ATP, is consistent with the involvement of a G protein-coupled receptor to which category P2Y but not P2X receptors belong.
To further support the hypothesis that ATP-stimulated prostacyclin release is mediated by P2 and not P1 receptors, we attempted to stimulate the release of of prostacyclin using adenosine. Consistent with others (2, 4, 20, 25), adenosine failed to stimulate the release of prostacyclin from these veins. To ensure that the response was not mediated via AMP, which also exerts its effect via stimulation of P1 receptors, we tested the nonselective P1 antagonist XAC on ATP-stimulated prostacyclin release. This antagonist did not alter the ATP-mediated response. Only one concentration of XAC was utilized in the present study so it is possible that a higher concentration may have had an effect. However, this concentration has been shown to effectively inhibit arteriolar dilations to 100 µM adenosine and to inhibit arteriolar dilations induced by muscle contraction (6, 19). These results are consistent with those of others who have demonstrated that neither adenosine nor AMP stimulate the release of prostacyclin from endothelial cells (4, 20, 25).
The results obtained from the present study support our hypothesis that ATP stimulates the release of prostacyclin from venous endothelial cells via activation of P2Y receptors. Neither AMP nor adenosine is likely to be involved in this response; however, we cannot rule out the possibility that ADP, which, like ATP, exerts its effects via P2Y receptors, is playing a role. These results support a role for the venous endothelium in control of vascular tone.
Perspectives
Skeletal muscle blood flow is regulated to meet the metabolic needs of the tissue. The vasculature is arranged as a series of successively branching arterioles, all of which must dilate during periods of increased muscle metabolism. We have hypothesized that the close arteriolar-venular pairing allows the arteriolar diameter to be influenced by vasoactive substances diffusing from the venules. The venular side of the circulation undergoes dramatic changes in blood chemistry during increased muscle metabolism and is thus in an optimal position to monitor the overall metabolic state of the tissue. Red blood cells can release ATP in response to changes in blood chemistry and may provide the means of communication between the metabolic state of the tissue and the vasculature. ATP-stimulated release of vasodilators, such as prostacyclin, from venules would ensure that larger arterioles located beyond the site of increased muscle metabolism and possibly beyond the distance which a conducted response can be transmitted would dilate rapidly and thus contribute to the increase in blood supply to the exercising muscle.
|
ACKNOWLEDGMENTS |
|---|
This work was supported by grants from the National Institutes of Health (HL-51971, HL 63958) and the American Heart Association (National Grant-in-Aid and Southeast Affiliate Postdoctoral Fellowship).
|
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 |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  M. Kavdia and A. S. Popel Venular endothelium-derived NO can affect paired arteriole: a computational model Am J Physiol Heart Circ Physiol, February 1, 2006; 290(2): H716 - H723. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
