(Hypertension. 2001;38:833.)
© 2001 American Heart Association, Inc.
Scientific Contributions |
From the Centre for Clinical Pharmacology, University College London, The Rayne Institute, London, United Kingdom.
Correspondence to Dr Amrita Ahluwalia, Centre for Clinical Pharmacology, University College London, The Rayne Institute, 5 University St, London, United Kingdom WC1E 6JJ. E-mail a.ahluwalia{at}ucl.ac.uk
| Abstract |
|---|
|
|
|---|
200 µm) were isolated and mounted in a pressure myograph. Internal diameter was recorded over a pressure range of 10 to 80 mm Hg. Removal of the endothelium significantly (P<0.05) enhanced the magnitude of myogenic constriction in wild-type mice. Similarly, pretreatment of arteries with NG-nitro-L-arginine methyl ester (L-NAME; 300 µmol/L) produced a comparable significant (P<0.05) increase in myogenic tone, whereas indomethacin (5 µmol/L) had no effect. eNOS knockout arteries also exhibited myogenic constriction. Neither L-NAME nor indomethacin had any effect on myogenic tone in the arteries of eNOS knockout mice. However, blockade of potential endothelium-derived hyperpolarizing factorlike mechanisms via inhibition of K+ flux using either apamin (100 nmol/L) with charybdotoxin (100 nmol/L), Ba2+ (30 µmol/L) with ouabain (1 mmol/L), or 18
-glycyrrhetinic acid (100 µmol/L) significantly (P<0.01) enhanced myogenic constriction. This study demonstrates that basal endothelium-derived NO modulates myogenic tone in mesenteric small arteries of wild-type mice. However, eNOS knockout arteries display normal myogenic responsiveness despite the absence of basal NO activity. The data suggest that this compensatory effect is due to the activity of an endothelium-derived hyperpolarizing factor to normalize vascular tone.
Key Words: nitric oxide endothelium mice nitric oxide synthase autoregulation
| Introduction |
|---|
|
|
|---|
Numerous studies have excluded a role for the endothelium in mediation of the myogenic response (see Meininger and Davis6). Recently, however, there has been renewed interest in the endothelium as a modulator rather than as a mediator of myogenic constriction.6 Studies of mesenteric vessels from spontaneously hypertensive rats suggest that endothelium removal enhances myogenic constriction,7 possibly as a consequence of the loss of vasodilator endothelium-derived NO. In addition, NO synthase (NOS) inhibitors potentiate pressure-induced constriction of large, but not small, hamster cremaster arterioles in vivo.8 Similarly, in the presence of estrogen, NOS inhibition modulates myogenic constriction of rat arteries (see Wellman et al9). The aim of the present study was to investigate the role of NO in pressure-mediated responses in small vessels to test the hypothesis that endothelial NOS (eNOS) activity is involved in the modulation of myogenic responses in isolated resistance arteries. Because highly specific inhibitors for each NOS isoform10 are not yet available, we conducted experiments in vessels from mice with targeted disruption of the eNOS gene to assess the importance of this particular NOS isoform.
| Methods |
|---|
|
|
|---|
Studies of Pressure-Mediated Responses In Vitro
Vessels were mounted in a 10-mL perfusion myograph chamber onto 2 opposing PSS-filled cannulas. The chamber was continuously superfused with PSS at 10 mL/min (37°C, pH 7.4), gassed with 21% O2/5% CO2 in N2, and placed on the stage of an inverted microscope (Nikon, TMS). The vessel was visualized using a video camera (VM-902; Hitachi Denshi Ltd), and the internal diameter was determined using a video dimension analyzer (Living Systems Inc)12 and recorded on a pen chart recorder (BS-272; Gould Electronics Ltd). After equilibration, vessels were pressurized to 80 mm Hg; those that did not develop spontaneous myogenic tone were rejected. Endothelial integrity was tested using acetylcholine (10 µmol/L); vessels not demonstrating >50% reversal of myogenic tone in response to acetylcholine were rejected. In vessels from eNOS knockout mice, acetylcholine does not always produce sufficient reversal of tone13; in these instances, the endothelium-dependent vasodilator bradykinin (1 µmol/L) was tested. Reversal of myogenic tone of >50% was taken as indicative of intact endothelium.
Pressure-diameter curves were constructed under no-flow conditions in the absence and then in the presence of drug treatment, Ca2+-free PSS containing 2 mmol/L EGTA, or endothelium denudation. At the conclusion of each experiment, vessels were bathed in Ca2+-free PSS containing 2 mmol/L EGTA at 80 mm Hg to determine passive diameter. Endothelium removal was achieved by the injection of air.14 U46619 (11
,9
-epoxymethano-PGH2, 10 nmol/L) was used to elevate tone, and acetylcholine (10 µmol/L) was used to test endothelium integrity. Vessels were considered endothelium denuded when acetylcholine reversal of tone was <10%. Smooth muscle sensitivity was tested using S-nitroso-N-acetyl penicillamine (SNAP; 10 µmol/L).
The NOS and cyclooxygenase inhibitors NG-nitro-L-arginine methyl ester (L-NAME; 300 µmol/L) and indomethacin (5 µmol/L) were used to determine the roles of NO and prostanoids, respectively. Involvement of endothelium-derived hyperpolarizing factor (EDHF) was determined using apamin (100 nmol/L) and charybdotoxin (100 nmol/L),15 the gap junction inhibitor 18
-glycyrrhetinic acid (100 µmol/L),16 or Ba2+ (30 µmol/L) and ouabain (1 mmol/L).17 Apamin and charybdotoxin in PSS were perfused (10 µL/min) through the artery and left in vessel contact for 30 minutes. All other agents were superfused for 30 minutes before construction of the second pressure curve.
Studies of Vascular Reactivity to NO Donors
Arteries were mounted in an automated isometric tension myograph (JP Trading) and bathed in PSS at 37°C.18 After equilibration19 in U-46619 (0.03 to 1 µmol/L)precontracted vessels, relaxation concentration-response curves to SNAP (1 nmol/L to 30 µmol/L) were constructed in the absence or presence of charybdotoxin and apamin or Ba2+ and ouabain.
Data Analysis
All values are expressed as the arithmetic mean±SEM. Statistical analysis was performed by paired or unpaired Students t test or ANOVA followed by Bonferronis correction for multiple comparisons. Differences were considered significant at P<0.05.
An expanded Methods section can be found in an online data supplement available at http://www.hypertensionaha.org.
Materials
U46619 was purchased from BIOMOL, and all other drugs were from Sigma Chemical Co.
| Results |
|---|
|
|
|---|
|
|
Endothelium-Derived NO Moderates Myogenic Responses in Wild-Type Arteries
Endothelium removal significantly (n=6, P<0.05) increased myogenic tone (Figure 2A) and abolished dilator responses to acetylcholine without affecting the response to U46619 or SNAP (Table). L-NAME treatment (n=6) also significantly (P<0.05) enhanced myogenic tone (Figure 2B), whereas indomethacin (n=6) (Figure 2C) had no effect.
|
K+ Channel Blockers Have Little Effect on Myogenic Constriction of Wild-Type Arteries
Charybdotoxin and apamin treatment enhanced myogenic constriction only at 70 mm Hg (n=5, P<0.05) in wild-type arteries. This drug combination did not alter U-46619induced constriction (49.3±2.9% in the absence and 51.2±5.3% in the presence) but attenuated acetylcholine-induced dilatation (63.7±14.8% in the absence and 41.8±15.8% in the presence, P<0.05) and relaxation to SNAP (pEC50 7.4±0.2 in the absence and 6.9±0.3 in the presence, n=5, P<0.001). In contrast, ouabain and Ba2+ (n=4) or 18
-glycyrrhetinic acid (n=4) did not alter myogenic responses (Figures 2E and 2F) or relaxation to SNAP (n=5).
Inhibitors of EDHF Responses Increase Myogenic Constriction of eNOS Knockout Arteries
Myogenic tone of eNOS knockout arteries was not altered by NOS (n=5, Figure 3A) or cyclooxygenase (n=5, Figure 3B) inhibition. However, myogenic tone was significantly increased in the presence of charybdotoxin and apamin (n=4, P<0.01, Figure 3C), Ba2+ and ouabain (n=5, P<0.01, Figure 3D), or 18
-glycyrrhetinic acid (n=4, P<0.05, Figure 3E). 18
-Glycyrrhetinic acid also abolished acetylcholine dilatation (47.8±12.0% in the absence and 0% in the presence) without altering U-46619induced constriction (41.0±11% and 35.5±6.6%, respectively) in eNOS knockout arteries.
|
| Discussion |
|---|
|
|
|---|
Third-order mesenteric small arteries from eNOS wild-type mice exhibited a typical pressure-dependent decrease in diameter within a physiologically relevant pressure range. Normal in vivo intraluminal pressure in arteries of the size used in the present study (ie,
100 µm) is thought to be 60% to 80% of mean arterial pressure (see Mulvany and Aalkjaer18 for a review). Therefore, with a mean arterial pressure of 95 to 100 mm Hg, intraluminal pressure of these arteries should range from 57 to 80 mm Hg (ie, within the range tested). The myogenic constrictor response first described by Bayliss in 1902 is considered to involve influx of extracellular Ca2+,5 and consistent with this hypothesis, myogenic tone was abolished in the absence of extracellular Ca2+. The pressure-passive diameter curve (Figure 1B) demonstrates that the vessels that were studied develop myogenic tone at pressures of >40 mm Hg, which is similar to responses in vessels of most other species, including humans.20 The presence of myogenic constriction is associated with a plateau in wall tension consistent with the suggestion that wall tension may be the stimulus and regulated variable for myogenic constriction.1
Small arteries of eNOS knockout mice also exhibited pressure-dependent constriction that is dependent on the presence of extracellular Ca2+. However, the absolute magnitude of this response was less than that of wild-type arteries. This did not appear to be due to any gross abnormalities in vessel wall structure consequent to eNOS deletion, because no differences were evident between the 2 strains in basal and passive diameters or wall thickness. It is unlikely that the difference seen was an artifact due to the pressure range used, because the mean arterial pressure of the eNOS knockout is
110 mm Hg,11 and therefore the physiological pressure range for small arteries in these animals falls well within the intraluminal pressures used. An alternative explanation for the apparent suppression in myogenic tonus may be that in the absence of eNOS, other dilator factors are upregulated, such that an overcompensation for the loss of basal NO activity is produced. Indeed, this appears to be the case in the present study.
The importance of the endothelium was assessed by construction of pressure-diameter curves before and after endothelium removal, achieved by bolus injection of air through the vessel. The vessel was considered denuded if dilatation to the endothelium-dependent dilator acetylcholine was lost. This procedure did not damage the underlying smooth muscle because vasoconstriction to U46619 and dilation to the endothelium-independent dilator SNAP were unaffected. Removal of the endothelium from eNOS wild-type arteries greatly increased the magnitude of myogenic constriction. These findings are in agreement with a recent study in mesenteric small arteries of spontaneously hypertensive rats in which myogenic tone was greater in the absence of the endothelium.21 In contrast, studies in a variety of other vessel types show no effect of endothelium removal on the pressure-diameter relationship.6 The difference between these studies and the present study may relate to the method of endothelium removal that was used; many of the studies used chemical rather than physical removal procedures. Alternatively, the differences may reflect true differences in endothelial regulation of myogenic tone between species or vascular beds.
NO is an important mediator of vascular autoregulation. In isolated resistance arteries, changes in flow (shear stress) result in eNOS-derived NO release from the endothelium to produce dilatation and thereby normalize shear stress.14,22 In the present study, we demonstrated that NO is also involved in transmural pressure-mediated responses. Inhibition of NO synthesis in eNOS wild-type arteries significantly decreased intraluminal diameter only in the pressure range at which myogenic tone was observed (ie, >40 mm Hg). The lack of effect of L-NAME at intraluminal pressures of <50 mm Hg indicates that the extent of basal NO release has minimal impact on vascular tone of these vessels at low pressures. In keeping with previous results,9 the increase in magnitude of the myogenic response after NOS inhibition was similar to the effect of endothelium removal. Inhibition of cyclooxygenase, using indomethacin, had no effect on myogenic tone. Together, these findings show that the endothelium functionally moderates myogenic tone and that the endothelium-derived factor responsible for this effect is NO.
Because NOS inhibitors show little selectivity for the different NOS isoforms, we used mice with targeted disruption of the eNOS gene to determine whether modulation of myogenic constriction was provided by eNOS activity. We focused on eNOS because it is the main isoform found in the endothelium. Unlike mesenteric eNOS wild-type arteries, NOS inhibition had no effect on the pressure-diameter relationship of eNOS knockout arteries. However, contrary to expectation, the magnitude of myogenic tone in these vessels was not greater than that observed in wild-type arteries, and if anything, it was less. These results imply developmental adaptation and, possibly, overcompensation for the loss of endothelium-derived NO. Indeed, other studies demonstrate similar adaptation in other important physiological responses in eNOS knockout animals. For example, the absence of eNOS is compensated for by neuronal NOS in pial arteries23 and by cyclooxygenase in mesenteric arteries,24 both in the response to acetylcholine25 and in flow-mediated responses in gracilis muscle arterioles.26 However, in the present study, neither nNOS nor cyclooxygenase compensated for the loss of eNOS, because L-NAME and indomethacin had no effect.
An alternative endothelium-derived mediator that might compensate for the loss of NO is EDHF. Similar to NO and prostacyclin, EDHF is another prominent vasodilator factor released from the endothelium in response to endothelium-dependent vasodilators such as acetylcholine and bradykinin. EDHF causes vasodilatation by hyperpolarizing the adjacent smooth muscle. In addition, studies suggest that the role of EDHF in regulation of arterial vascular tone increases as the diameter of the blood vessel decreases, such that EDHF has little role in mediating endothelium-dependent vasodilatation of conduit arteries but a major role in resistance arteries.27 Although it is clear that EDHF release and activity involve alteration in K+ flux of both the endothelial and smooth muscle cell, the exact mechanisms involved are uncertain and highly controversial. Currently, the clearest method for identifying the involvement of EDHF can be achieved through the combined inhibition of large and small conductance Ca2+-activated K+ channels (using charybdotoxin and apamin,15 respectively). In the present study, replacement of the intraluminal solution with PSS containing charybdotoxin and apamin profoundly enhanced myogenic constriction in eNOS knockout arteries at pressures of >40 mm Hg. There is some suggestion that large conductance Ca2+-activated K+ channels (BKCa) determine the extent of myogenic tone per se,26 and therefore the increase in tone may not have been due solely to the inhibition of EDHF. However, in the present study, the channel blockers were administered intraluminally rather than the abluminally, and therefore the effects seen are likely to be due primarily to an effect on endothelial cells rather than on smooth muscle cells. Indeed, previous studies in rat mesenteric small arteries demonstrate that EDHF-mediated relaxation to acetylcholine is inhibited by charybdotoxin and apamin only when selectively applied to the lumen.28 Furthermore, these inhibitors do not affect myogenic constriction or U-46619induced tone when applied abluminally,28 supporting our conclusion that the effect seen in the present study was due to inhibition of an EDHF response. The combination of charybdotoxin and apamin also enhanced myogenic constriction of wild-type arteries, suggesting that at least part of the response even in the presence of eNOS may be due to hyperpolarization.29 However, in wild-type arteries, the effects of these inhibitors were far less than those seen in eNOS knockout arteries. In wild-type arteries, charybdotoxin and apamin partially blocked the responses to NO, as shown by the suppression of SNAP dilatation in the presence of these inhibitors, and this may explain the efficacy of these inhibitors against myogenic tone.
To provide additional support for a role of EDHF as a moderator of myogenic constriction in eNOS knockout arteries, we studied other proposed inhibitors of EDHF. Inhibition of inwardly rectifying K+ channels (Kir) with Ba2+ and Na+,K+-ATPase with ouabain blocks EDHF activity at the smooth muscle cell in certain vascular preparations.15 In the present study, this combination of inhibitors moderately enhanced myogenic tone only in eNOS knockout arteries and not in eNOS wild-type arteries. This suggests that Kir and Na+,K+-ATPase may play a small role in mediation of the EDHF component in eNOS knockout arteries. A similar result was obtained using a selective blocker of gap junctions (18
-glycyrrhetinic acid). Recent theory suggests that endothelium-dependent hyperpolarization is consequent to passage, from the endothelial cell to the smooth muscle cell, of a factor or electrotonic current via myoendothelial gap junctions.30 Similar to the other inhibitors of EDHF, 18
-glycyrrhetinic acid also enhanced myogenic constriction of eNOS knockout arteries, with no effect in wild-type arteries, but as with Ba2+ and ouabain, the effect was moderate. Together, the results suggest that hyperpolarization in part mediates endothelium-induced modulation of myogenic constriction in both wild-type and eNOS knockout arteries but that the factor or mechanism responsible for this effect is different in the 2 species (ie, NO in wild-type and EDHF in eNOS knockout arteries). This EDHF component involves activation of large and small conductance Ca2+-dependent K+ channels with a minor role for Kir and Na+,K+-ATPase and myoendothelial gap junctions. Ideally, measurement of membrane potential demonstrating hyperpolarization or rather repolarization in response to increases in intraluminal pressure would provide additional support for a role for an EDHF. However, these experiments are technically difficult because elevation of pressure results in diameter changes that would dislodge an electrode impaled in the vessel wall, thus making sustained measurement of membrane potential in a single smooth muscle cell impossible. Regardless of this, the data suggest that where eNOS activity is compromised or absent, EDHF is upregulated as a compensatory mechanism to preserve endothelium-derived vasodilator regulation of vascular tone. In support of such a back-up mechanism are findings that demonstrate basal eNOS-derived NO inhibition of EDHF release.31 In the present study, however, it is unlikely that this is an acute back-up mechanism because in arteries of wild-type mice, NOS inhibition produced a pronounced increase in myogenic tone: if NO-induced intrinsic inhibition could be alleviated acutely, myogenic constriction would be expected to remain relatively unchanged.
Finally, we attempted to investigate the effects of endothelium removal on myogenic responses of arteries from eNOS knockout animals. However, using the same protocol as that used for wild-type arteries, it was not possible to remove the endothelium, despite reapplication of air up to 3 additional times. Similar attempts to remove the endothelium with detergent, such as Triton X or CHAPS, also failed, as demonstrated by the fact that endothelium-dependent dilators (acetylcholine and bradykinin) still caused relaxation. The reason for the difficulty in removing the endothelium in microvessels of eNOS knockout arteries is unclear and may be worthy of investigation in its own right.
In conclusion, we have demonstrated that in the murine mesenteric microcirculation, NO synthesized by eNOS in the endothelium significantly modulates Bayliss-type autoregulation. Therefore, eNOS-derived NO release may provide an important negative feedback mechanism in autoregulation of resistance arteries, and this has implications for understanding how the loss of NO in disease states might affect tissue and organ perfusion as well as blood pressure responses. Moreover, in the chronic absence of eNOS-derived NO, an EDHF-like factor or factors compensate for the lack of NO such that a vasodilator influence over myogenic constriction of resistance arteries is maintained. It remains to be determined whether these findings account for the observation that EDHF-like responses are more prominent in humans with hypertension in whom NO-mediated dilatation is impaired.32
| Acknowledgments |
|---|
Received November 7, 2000; first decision December 4, 2000; accepted March 23, 2001.
| References |
|---|
|
|
|---|
glycyrrhetinic acid. Br J Pharmacol. 1998; 125: 13.[Medline]
[Order article via Infotrieve]
This article has been cited by other articles:
![]() |
R. Schubert, D. Lidington, and S.-S. Bolz The emerging role of Ca2+ sensitivity regulation in promoting myogenic vasoconstriction Cardiovasc Res, January 1, 2008; 77(1): 8 - 18. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. M. Fitzgerald, B. K. Kemp-Harper, H. C. Parkington, G. A. Head, and R. G. Evans Endothelial dysfunction and arterial pressure regulation during early diabetes in mice: roles for nitric oxide and endothelium-derived hyperpolarizing factor Am J Physiol Regulatory Integrative Comp Physiol, August 1, 2007; 293(2): R707 - R713. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. Adlam, J. K. Bendall, J. P. De Bono, N. J. Alp, J. Khoo, T. Nicoli, M. Yokoyama, S. Kawashima, and K. M. Channon Cardiovascular Control: Relationships between nitric oxide-mediated endothelial function, eNOS coupling and blood pressure revealed by eNOS-GTP cyclohydrolase 1 double transgenic mice Exp Physiol, January 1, 2007; 92(1): 119 - 126. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. G. Hemmings, S. J. Williams, and S. T. Davidge Increased myogenic tone in 7-month-old adult male but not female offspring from rat dams exposed to hypoxia during pregnancy Am J Physiol Heart Circ Physiol, August 1, 2005; 289(2): H674 - H682. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. G. Hemmings, S. Veerareddy, P. N. Baker, and S. T. Davidge Increased Myogenic Responses in Uterine but not Mesenteric Arteries from Pregnant Offspring of Diet-Restricted Rat Dams Biol Reprod, April 1, 2005; 72(4): 997 - 1003. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Bucci, F. Roviezzo, I. Posadas, J. Yu, L. Parente, W. C. Sessa, L. J. Ignarro, and G. Cirino Endothelial nitric oxide synthase activation is critical for vascular leakage during acute inflammation in vivo PNAS, January 18, 2005; 102(3): 904 - 908. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. M. Bauer, J. Yu, Y. Chen, R. Hickey, P. N. Bernatchez, R. Looft-Wilson, Y. Huang, F. Giordano, R. V. Stan, and W. C. Sessa Endothelial-specific expression of caveolin-1 impairs microvascular permeability and angiogenesis PNAS, January 4, 2005; 102(1): 204 - 209. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. S. Scotland, S. Chauhan, C. Davis, C. De Felipe, S. Hunt, J. Kabir, P. Kotsonis, U. Oh, and A. Ahluwalia Vanilloid Receptor TRPV1, Sensory C-Fibers, and Vascular Autoregulation: A Novel Mechanism Involved in Myogenic Constriction Circ. Res., November 12, 2004; 95(10): 1027 - 1034. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Veerareddy, C.-L. M. Cooke, P. N. Baker, and S. T. Davidge Gender differences in myogenic tone in superoxide dismutase knockout mouse: animal model of oxidative stress Am J Physiol Heart Circ Physiol, July 1, 2004; 287(1): H40 - H45. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. Gros, T. Afroze, X.-M. You, G. Kabir, R. Van Wert, W. Kalair, A. E. Hoque, I. N. Mungrue, and M. Husain Plasma Membrane Calcium ATPase Overexpression in Arterial Smooth Muscle Increases Vasomotor Responsiveness and Blood Pressure Circ. Res., October 3, 2003; 93(7): 614 - 621. [Abstract] [Full Text] [PDF] |
||||
![]() |
Z. L. S. Brookes and S. Kaufman Myogenic responses and compliance of mesenteric and splenic vasculature in the rat Am J Physiol Regulatory Integrative Comp Physiol, June 1, 2003; 284(6): R1604 - R1610. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. A. Ortiz and J. L. Garvin Cardiovascular and renal control in NOS-deficient mouse models Am J Physiol Regulatory Integrative Comp Physiol, March 1, 2003; 284(3): R628 - R638. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Veerareddy, C.-L. M. Cooke, P. N. Baker, and S. T. Davidge Vascular adaptations to pregnancy in mice: effects on myogenic tone Am J Physiol Heart Circ Physiol, December 1, 2002; 283(6): H2226 - H2233. [Abstract] [Full Text] [PDF] |
||||
![]() |
Y. Iwakiri, G. Cadelina, W. C. Sessa, and R. J. Groszmann Mice with targeted deletion of eNOS develop hyperdynamic circulation associated with portal hypertension Am J Physiol Gastrointest Liver Physiol, November 1, 2002; 283(5): G1074 - G1081. [Abstract] [Full Text] [PDF] |
||||
![]() |
L. B. Tanko and K. Matrougui Can We Apply Results From Large to Small Arteries? Circ. Res., March 22, 2002; 90 (5): e68 - e68. [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Hypertension Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 2001 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |