This Article Abstract Full Text (PDF) All Versions of this Article: 92/2/151    most recent 01.RES.0000052671.53256.49v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Miura, H. Articles by Gutterman, D. D. Search for Related Content PubMed PubMed Citation Articles by Miura, H. Articles by Gutterman, D. D. Related Collections Coronary circulation Other diabetes Other Vascular biology

(Circulation Research. 2003;92:151.)
© 2003 American Heart Association, Inc.

Clinical Research

Diabetes Mellitus Impairs Vasodilation to Hypoxia in Human Coronary Arterioles

Reduced Activity of ATP-Sensitive Potassium Channels

Hiroto Miura, Ruth E. Wachtel, Fausto R. Loberiza, Jr, Takashi Saito, Mamoru Miura, Alfred C. Nicolosi, David D. Gutterman

From the Department of Veterans Affairs Medical Center (H.M., D.D.G.), Milwaukee, Wis; Department of Veterans Affairs Medical Center (R.E.W.) and Department of Anesthesia, University of Iowa College of Medicine (R.E.W.), Iowa City, Iowa; Department of Medicine and Cardiovascular Research Center (H.M., D.D.G.), Health Policy Institute (F.R.L.), and Department of Surgery (A.C.N.), Medical College of Wisconsin, Milwaukee, Wis; and 2nd Department of Internal Medicine (T.S., M.M.), Akita University, Akita City, Japan.

Correspondence to Hiroto Miura, Department of Medicine and Cardiovascular Research Center, Medical College of Wisconsin, 8701 Watertown Plank Rd, Milwaukee, WI 53226. E-mail hmiura{at}mcw.edu

	Abstract

Top Abstract Introduction Materials and Methods Results Discussion References

 
ATP-sensitive K⁺ channels (K_ATP) contribute to vasomotor regulation in some species. It is not fully understood the extent to which K_ATP participate in regulating vasomotor tone under physiological and pathophysiological conditions in the human heart. Arterioles dissected from right atrial appendage were studied with video microscopy, membrane potential recordings, reverse transcription–polymerase chain reaction, and immunohistochemistry. Hypoxia produced endothelium-independent vasodilation and membrane hyperpolarization of vascular smooth muscle cells, both of which were attenuated by glibenclamide. Aprikalim, a selective K_ATP opener, also induced a potent endothelium-independent and glibenclamide-sensitive vasodilation with membrane hyperpolarization. Reverse transcription–polymerase chain reaction detected mRNA expression for K_ATP subunits, and immunohistochemistry confirmed the localization of the inwardly rectifying Kir6.1 protein in the vasculature. In patients with type 1 or type 2 diabetes mellitus (DM), vasodilation was reduced to both aprikalim (maximum dilation, DM(+) 90±2% versus DM(-) 96±1%, P<0.05) and hypoxia (maximum dilation, DM(+) 56±8% versus DM(-) 85±5%, P<0.01) but was not altered to sodium nitroprusside or bradykinin. Baseline myogenic tone and resting membrane potential were not affected by DM. We conclude that DM impairs human coronary arteriolar dilation to K_ATP opening, leading to reduced dilation to hypoxia. This reduction in K_ATP function could contribute to the greater cardiovascular mortality and morbidity in DM.

Key Words: human • coronary microcirculation • ATP-sensitive potassium channels • hypoxia • diabetes mellitus

	Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

 
ATP-sensitive K⁺ channels (K_ATP) are present in a variety of tissues. K_ATP participate in vasodilation to hypoxia and ischemia.^1,2 K_ATP function is altered in the presence of diabetes mellitus (DM). In some species, vasodilation to K_ATP openers is impaired by DM,^3–5 whereas in other species and vascular beds, the vasodilation may be augmented.⁶ We examined whether K_ATP opening elicits coronary arteriolar dilation in humans and whether K_ATP are responsible for hypoxia-induced vasodilation. We also determined whether K_ATP-mediated vasodilation is impaired in coronary vessels from subjects with DM. This could have important implications for patients with DM, who may suffer from reduced metabolic vasodilation.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

 
Materials
Aprikalim (RP52891, Rhone-Poulenc Rorer, France) was prepared in 50% ethanol and distilled water. Endothelin-1 (Peninsula Laboratories, Inc) was prepared in saline with 1% BSA. All other chemicals were obtained from Sigma Chemical Co and dissolved in distilled water, except indomethacin, which was dissolved in distilled water with 1 N NaOH, and the pH was adjusted to 7.4 with 0.1 N HCl. All concentrations represent the final molar concentrations (mol/L) in the organ chambers.

General Preparation
Fresh specimens of right atrial appendage were obtained from 200 patients undergoing cardiac surgery. All protocols and procedures were approved by the appropriate institutional review boards. After surgical removal, the atrial appendage was placed in cold oxygenated Krebs solution, as described previously.^7–11 Human coronary arterioles (HCAs) were isolated and transferred to an organ chamber containing oxygenated warmed (37°C) Krebs solution, where they were connected via glass micropipettes to a hydrostatic reservoir. The preparation was transferred to the stage of an inverted microscope (CK2, Olympus) for continuous measurement of internal diameter.^7–11 All pharmacological agents were added to the external bathing solution. One vessel was used from each subject.

HCAs that failed to constrict by >30% of expected passive diameter to 75 mmol/L KCl were discarded. All HCAs dilated to adenosine diphosphate (ADP, 10^-5 mol/L), confirming integrity of endothelial function.⁷ Endothelin-1 (10^-10 mol/L) or acetylcholine (5x10^-7 mol/L)^7–9,11,12 was, if needed, added to adjust basal tone to a consistent level between 30% to 60% of passive diameter.

Aprikalim Treatment
After constriction, vasodilation to cumulative increases in the concentration of aprikalim (10^-10 to 10^-5 mol/L, a selective K_ATP opener) was studied in the presence and absence of N-nitro-L-arginine methyl ester (L-NAME) (10^-4 mol/L, an NO synthase inhibitor), indomethacin (10^-5 mol/L, a cyclooxygenase inhibitor), glibenclamide (10^-6 mol/L, a selective K_ATP blocker), or mechanical endothelial denudation.^7–11 Reversibility of dilation to aprikalim was tested after a 30-minute exposure to and after washout of glibenclamide. Bradykinin (BK) (10^-12 to 10^-6 mol/L), an endothelium-dependent vasodilator, was tested in the presence of L-NAME and indomethacin.⁸ Vasodilation to sodium nitroprusside (SNP) (10^-10 to 10^-4 mol/L) or papaverine (10^-4 mol/L), endothelium-independent dilators, was also tested. In some vessels, diazoxide (10^-8 to 10^-4 mol/L, another K_ATP opener) was also used.

Hypoxia Induction
Hypoxic conditions were produced by bubbling Krebs solution with 5% CO₂ and 95% N₂ gas in a covered vessel chamber. Vascular responses to two consecutive 15-minute periods of hypoxia separated by 60 minutes of intervening normoxia were examined in the absence and presence of inhibitors or denudation. Denudation was confirmed by a reduction in dilation to ADP (10^-5 mol/L) to <10% (after -4±7% versus before 85±6%, P<0.05) and preserved dilation to SNP (10^-4 mol/L; after 97±2% versus before 99±1%, P=NS).^7,11 Responses to aprikalim were also studied before and after hypoxia.

Measurement of Vascular Smooth Muscle Membrane Potential
We measured resting membrane potential (E_m) of vascular smooth muscle cells (VSMCs) and changes in E_m to aprikalim or hypoxia, as described previously.^8,9,13 Briefly, pressurized HCAs were impaled from the adventitial surface with a glass microelectrode (40 to 90 M impedance, filled with 3 mol/L KCl) and connected to a high-impedance biological amplifier (Axoclamp, Axon Instruments).

Reverse Transcription–Polymerase Chain Reaction Analysis
From each subject, 5 to 10 HCAs of 1 to 2 mm in length were isolated. Extraction and isolation of mRNA was achieved with oligo(dT)₂₅-linked magnetic beads (Dynal), and first-strand cDNA was synthesized from mRNA using Ready-To-Go You-Prime First-Strand Beads (Amersham Biosciences) according to established protocols.¹⁴ Reverse transcription was initiated by addition of 0.5 µg Oligo(dT)₁₅ Primer (Promega) at 37°C for 75 minutes. Protocols were performed using 1 µL of the cDNA in 30 µL reaction containing 0.2 µmol/L of each primer specific for K_ATP, 200 µmol/L of each dNTP, and 0.7 U of Taq polymerase (PCR Supermix, Life Technologies). Primers were used for detection of the inwardly rectifying Kir6.1 (expected product, 737 bp) and Kir6.2 (636 bp) and the sulfonylurea receptors SUR1 (134 bp) and SUR2A/2B (451/312 bp), as described before.^15,16 The primers for SUR2A/2B were designed to identify both SUR2A and SUR2B. Cycling conditions were as follows: for Kir6.1 and 6.2, 96°C, 3 minutes, 40 cycles (96°C, 15 seconds; 55°C, 30 seconds; 72°C, 78 seconds) and 72°C, 7 minutes; and for SUR1 and SUR2A/2B, 94°C, 3 minutes, 40 cycles (94°C, 30 seconds; 58.8°C, 1 minute; 72°C, 1 minute) and 72°C, 7 minutes. An aliquot (10 µL) of reverse transcription–polymerase chain reaction (RT-PCR) products was analyzed on 2% TEA agarose gels (Life Technologies). PCR products were visualized by staining with ethidium bromide with ultraviolet transillumination. In the experiments with no or faint expression, a second PCR was also conducted to confirm the results. Negative controls lacking cDNA or reverse transcriptase during cDNA synthesis did not amplify any products, indicating lack of contamination with genomic DNA.

Immunohistochemistry
Small pieces (1 mm³) of pectinate muscle were fixed with 4% paraformaldehyde in PBS, infiltrated with 20% sucrose HEPES buffer solution, and frozen in OCT compound. Sections (8-µm thick) were immunolabeled with a polyclonal antibody against Kir6.1 protein (dilution 1:250, Santa Cruz Biotechnology, Inc). Immunostains were visualized by using avidin-biotin horseradish peroxidase visualization systems (Vectastain Universal Quick kit, Vector Laboratories). As a control for nonspecific binding, the primary antibody was omitted.

Statistical Analysis
Vasodilations to aprikalim, hypoxia, and BK are expressed as percent dilation, with 100% representing the change in diameter to SNP (10^-4 mol/L). Vasodilation to SNP is expressed as percent maximal passive diameter obtained with Ca²⁺-free Krebs solution containing SNP (10^-4 mol/L). Student’s paired t test was used to compare responses to hypoxia, change in E_m, maximal dilations, and ED₅₀ values (the molar concentration of dilator that produced a 50% maximal response). Dilation to aprikalim under different conditions was compared using 2-way repeated-measures ANOVA with a Bonferroni correction. Multivariate analysis assessed the influence of age, sex, and underlying diseases (Table 1) on vasodilations, resting E_m, and vascular tone. Regression models for all doses or time intervals isolated the confounding effects of the other diseases on dilator responses. Analysis of covariance (ANACOVA) was performed to adjust for contributions of each factor to an impaired response. SAS for Windows, version 8, was used for analyses. Significance was defined as P<0.05. Data are expressed as mean±SEM.

View this table: [in this window] [in a new window]   Table 1. Demographics (n=200)

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

 
Two hundred HCAs with a passive internal diameter of 100±3 µm (range, 30 to 271 µm) were dissected. Patient demographics are summarized in Table 1.

Mechanism of Hypoxic Vasodilation
Hypoxia produced a gradual but potent vasodilation in HCAs (Figure 1; 25±6, 56±7, and 75±5% at 5, 10, and 15 minutes, respectively; n=29). The reduction in PO₂ was achieved within 5 minutes with minimal change in PCO₂ or pH. On reoxygenation, vascular tone recovered rapidly to resting levels after transient peak dilation at 1 minute (79±5%). PO₂, PCO₂, and pH returned to original levels within 5 minutes. No change in diameter was observed during normoxia of similar duration (n=9). Vessels exposed to two consecutive 15-minute periods of hypoxia separated by a 60-minute period of intervening normoxia demonstrated similar degrees of vasodilation (data not shown, P=NS, n=6).

View larger version (23K): [in this window] [in a new window]   Figure 1. Vascular response to hypoxia and reoxygenation. Hypoxia produced a prominent and reversible vasodilation of HCAs (n=29; male, 90%; age, 61±15 years; coronary artery disease, 93%; myocardial infarction, 7%; hypertension, 27%; hypercholesterolemia, 17%; DM, 30%; congestive heart failure, 7%). After reoxygenation, the diameter returned to original levels. No change in vascular diameter was observed during normoxic experiments of similar duration (n=9; male, 78%; age, 64±12 years; coronary artery disease, 89%; myocardial infarction, 22%; hypertension, 33%; hypercholesterolemia, 44%; DM, 44%; congestive heart failure, 22%).

Treatment with L-NAME did not alter vasodilation to hypoxia (75±7 versus control 87±9%, P=NS at 15 minutes, n=6). Indomethacin potentiated dilation only at 5 minutes (59±10 versus control 24±10%, P<0.05). Endothelial removal had no effect (68±10 versus control 75±14%, P=NS at 15 minutes, n=4). Thus, the endothelium is not critical for hypoxic vasodilation of HCAs.

Glibenclamide reduced hypoxia-induced vasodilation (Figure 2A; 27±6 versus control 65±7%, P<0.005 at 15 minutes, n=12). The reduced response to hypoxia observed in the presence of glibenclamide could conceivably be the result of impaired K_ATP function or an inability of K_ATP to respond to a second episode of hypoxia. Dilation to aprikalim was similar before and after exposure to hypoxia (data not shown, P=NS, n=8). Therefore, it is unlikely that hypoxia alters K_ATP activation during a subsequent period of hypoxia.

View larger version (16K): [in this window] [in a new window]   Figure 2. Role of KATP in dilation of HCAs to hypoxia and aprikalim. A, Glibenclamide (10-6 mol/L) reduced hypoxia-induced vasodilation (###P<0.005 vs control, n=12). Patient demographics: male, 100%; age, 67±9 years; coronary artery disease, 92%; myocardial infarction, 25%; hypertension, 58%; hypercholesterolemia, 17%; DM, 42%; congestive heart failure, 8%. B, Vessels dilated to aprikalim (10-10 to 10-5 mol/L) in a glibenclamide-sensitive (10-6 mol/L) manner (#P<0.05 vs control, n=5). Patient demographics: male, 60%; age, 51±25 years, coronary artery disease, 80%; myocardial infarction, 20%; hypertension, 0%; hypercholesterolemia, 0%; DM, 20%; congestive heart failure, 40%.

K_ATP-Mediated Vasodilation
We next tested whether K_ATP activation can dilate HCAs. Aprikalim produced potent vasodilation in a concentration-dependent manner with -log[ED₅₀] value of 7.0±0.2 (vehicle produced no dilation; data not shown). Glibenclamide inhibited dilation to aprikalim (Figure 2B; maximum dilation, 22±6 versus control 90±4%, P<0.05, n=5) and attenuated diazoxide-induced dilation (maximum dilation, 13±5 versus control 44±4%, P<0.05, n=6) but had no effect on dilation to SNP (maximum dilation, 86±5 versus control 87±6%, P=NS, n=6) or papaverine (maximum dilation, 95±2 versus control 92±4%, P=NS, n=6). We previously showed that this dose of glibenclamide has no effect on the dilation to BK⁸ or shear stress.¹⁰ Aprikalim-induced dilation was unchanged by L-NAME (-log[ED₅₀], 7.1±0.4 versus control 7.0±0.5, P=NS; maximum dilation, 93±3 versus control 93±3%, P=NS, n=5), indomethacin (-log[ED₅₀], 6.9±0.4 versus control 7.0±0.5, P=NS; maximum dilation, 87±6 versus control 93±3%, P=NS, n=5), or endothelial denudation (-log[ED₅₀], 6.4±0.3 versus control 6.2±0.3, P=NS; maximum dilation, 96±2 versus control 93±4%, P=NS, n=4). Thus, dilation to selective K_ATP opening with aprikalim is independent of the endothelium in HCAs.

Membrane Hyperpolarization to Aprikalim and Hypoxia
We measured the change in VSMCs E_m to K_ATP activation. Figure 3A shows a typical example of the effect of aprikalim on E_m of a VSMC in a cannulated and pressurized HCAs. Aprikalim produced hyperpolarization from a resting E_m of -52 to -54, -58, and -85 mV at 10^-9, 10^-7, and 10^-5 mol/L, respectively, in a dose-dependent fashion. BK did not additionally hyperpolarize the VSMC, indicating a near-maximal hyperpolarization. Increases in extraluminal KCl depolarized the VSMC. Summary data are shown in Figure 3B.

View larger version (17K): [in this window] [in a new window]   Figure 3. A, Changes in VSMC Em to aprikalim. Representative recording of Em in vessels at basal tone shows aprikalim-induced hyperpolarization. Arrows indicate time of drug application. Fifty-five-year-old man with coronary artery disease. B, Summary data from 6 vessels show that aprikalim hyperpolarizes VSMCs in a dose-dependent manner (#P<0.05 vs baseline, n=6). Patient demographics: male, 83%; age, 65±10 years; coronary artery disease, 67%; myocardial infarction, 0%; hypertension, 33%; hypercholesterolemia, 33%; DM, 17%; congestive heart failure, 0%. C, After membrane depolarization and vasoconstriction with acetylcholine, hypoxia induced hyperpolarization (#P<0.05 vs baseline, n=5) and vasodilation. Glibenclamide attenuated both vasodilation and hyperpolarization to hypoxia (*P<0.05 vs without glibenclamide, n=5). Patient demographics: male, 100%; age, 62±12 years; coronary artery disease, 80%; myocardial infarction, 20%; hypertension, 60%; hypercholesterolemia, 0%; DM, 40%; congestive heart failure, 20%.

To link vasodilator and electrophysiological observations, we simultaneously measured changes in vessel diameter and E_m during hypoxia. Figure 3C shows that hypoxia elicited dilation and hyperpolarization that were attenuated by glibenclamide (maximum dilation, 46±5 versus control 73±7%, P<0.05, n=5).

K_ATP Expression in the Coronary Microcirculation
In HCAs, RT-PCR detected transcripts for Kir6.1 and SUR2B from all subjects tested (Figures 4A and 4B), whereas mRNA for either Kir6.2, SUR1, or SUR2A was detected in only one subject.

View larger version (66K): [in this window] [in a new window]   Figure 4. Expression of KATP in the human coronary microcirculation. A, RT-PCR detected mRNA for Kir6.1 in all subjects tested, whereas mRNA for Kir6.2 was seen only in one subject. B, Left, transcripts for SUR1 were faintly detected in one of three vessels. Right, all subjects expressed mRNA for SUR2B, whereas one subject showed SUR2A mRNA expression. DNA size markers are indicated on the left, and each expected product size is on the right. Subject 1 was a 79-year-old woman with hypertension; subject 2 was a 65-year-old man; and subject 3 was a 55-year-old man. These patients had no coronary artery disease or DM. C, Representative photomicrograph of atrial tissue immunostained with a polyclonal antibody against Kir6.1. Strong positive staining (brown tint) was seen in VSMCs and modest staining in endothelial cells. Magnification x60 (scale bars=10 µm). Tissue was from a 47-year-old woman without coronary artery disease or DM.

Positive staining for Kir6.1 was observed by immunohistochemistry in HCAs (Figure 4C). VSMCs showed strong immunostaining for Kir6.1 protein, but endothelial cells were only faintly immunostained. This same pattern was observed in each of the three subjects.

Influence of Disease on K_ATP–Mediated Vasodilation
By multivariate analysis, only the presence of DM predicted the impaired vasodilation to hypoxia (Table 2). ANACOVA also indicated that DM was correlated with reduced dilation to hypoxia independent of other risk factors and conditions (Figure 5A; maximum dilation, DM(+) 56±8 versus DM(-) 85±5%, P<0.01, n=30).

View this table: [in this window] [in a new window]   Table 2. Multivariate Analysis of the Influence of Underlying Diseases, Sex, and Age on HCA Dilations

View larger version (24K): [in this window] [in a new window]   Figure 5. Vasodilation to hypoxia, aprikalim, SNP, and BK in DM adjusted for the other factors using ANACOVA based on the result of multivariate analysis. A, Hypoxia-induced vasodilation (n=30 [non-DM=21 and DM=9]) was reduced in the presence of DM compared with that in the absence of DM (##P<0.01 vs absence of DM). B, Presence of DM was associated with impaired dilation to aprikalim (n=34 [non-DM=23 and DM=11]) (#P<0.05, ##P<0.01, and ###P<0.005 vs absence of DM). C, SNP-induced vasodilation in DM adjusted for the influence of the other factors was comparable to that in non-DM (n=35 [non-DM=24 and DM=11]). D, Endothelium-derived hyperpolarizing factor–mediated vasodilation to BK (n=37 [non-DM=25 and DM=12]) was unchanged in the presence of DM.

DM was also identified by multivariate analysis and ANACOVA as the only independent predictor of reduced dilation to aprikalim (Table 2 and Figure 5B; -log[ED₅₀], DM(+) 6.3±0.3 versus DM(-) 7.3±0.2, P<0.05; maximum dilation, DM(+) 90±2 versus DM(-) 96±1%, P<0.05; n=34).

In contrast to hypoxia and aprikalim, no cardiovascular risk factors were associated with alterations in dilation to SNP or in endothelium-derived hyperpolarizing factor–mediated vasodilation to BK⁸ by multivariate analysis and ANACOVA (SNP, Figure 5C; -log[ED₅₀], DM(+) 7.1±0.3 versus DM(-) 7.1±0.2, P=NS; maximum dilation, DM(+) 97±2 versus DM(-) 92±1%, P=NS, n=35) and (BK, Figure 5D; -log[ED₅₀], DM(+) 7.6±0.2 versus DM(-) 7.6±0.1, P=NS; maximum dilation, DM(+) 92±5 versus DM(-) 90±3%, P=NS, n=37). Therefore, it is unlikely that impaired dilation to aprikalim or hypoxia in patients with DM is attributable to a nonspecific reduction in dilation.

Resting E_m from diabetic patients (n=14) and nondiabetic patients (n=35) was similar (DM -47±2 versus non-DM -47±1 mV, P=NS). No risk factor was predictive of an alteration in resting E_m by either multivariate analysis or ANACOVA. Spontaneous vascular tone was likewise not correlated with disease (data not shown).

Although we were not able to obtain full medication information, it is likely that some diabetic patients were treated with sulfonylureas, which may have influenced dilation to aprikalim. When diabetic patients were divided according to the subtype, hypoxia-induced vasodilation was similarly reduced in both type 1 (n=8) and type 2 (n=8) DM compared with non-DM controls (n=25) (Figure 6A; maximum dilation, type 1 21±3 and type 2 45±13 versus non-DM 81±5%, respectively, P<0.05). Vasodilation to aprikalim was similarly reduced in both type 1 (n=6) and type 2 (n=8) DM compared with the non-DM group (n=31) (Figure 6B; -log[ED₅₀], type 1 6.3±0.2 and type 2 6.1±0.2 versus non-DM 7.1±0.1, P<0.05 respectively; maximum dilation, type 1 84±12 and type 2 91±2 versus non-DM 95±1%, respectively, P=NS). In addition, aprikalim-induced dilation was unchanged before and after a 30-minute exposure and washout of glibenclamide (maximum dilation, after 97±2 versus before 99±0%; -log[ED₅₀], after 7.4±0.3 versus before 7.3±0.3, P=NS, n=6), indicating reversible inhibition of glibenclamide.

View larger version (23K): [in this window] [in a new window]   Figure 6. Comparison of dilator responses in patients with type 1 and type 2 DM. A, Hypoxia-induced dilation was impaired in both type 1 (n=8) and type 2 (n=8) DM patients (#P<0.05 vs non-DM, n=25). B, Aprikalim-induced dilation was also impaired in type 1 (n=6) and type 2 (n=8) DM patients (#P<0.05 vs non-DM, n=31).

DM could alter non–K_ATP-mediated mechanisms of hypoxic dilation. Glibenclamide reduced hypoxic vasodilation less in arterioles from subjects with DM (maximum dilation, 27±7 versus control 52±10%, P<0.05, n=6) than from subjects without DM (maximum dilation, 26±10 versus control 78±7%, P<0.05, n=6). However, the residual dilation after glibenclamide was similar, suggesting that the major impairment in hypoxic dilation in DM is attributable to the K_ATP-sensitive component.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

 
This study is the first to describe impaired vasodilation to hypoxia and K_ATP activation in HCAs from patients with DM. The major new findings are 5-fold. First, hypoxia induces a potent and endothelium-independent vasodilation. Second, K_ATP activation dose-dependently hyperpolarizes VSMCs and induces vasodilation. Third, the mechanism of hypoxia-induced vasodilation involves opening of K_ATP. Fourth, vasodilation to both hypoxia and K_ATP stimulation is impaired in both type 1 and type 2 DM. Finally, mRNA and protein for K_ATP subunits are present in HCAs, K_ATP possibly being comprised of SUR2B and Kir6.1. These findings indicate that in DM, vasodilator function associated with K_ATP opening is reduced in the human coronary microcirculation.

Vasodilation to Aprikalim
Human vessels can dilate in response to K_ATP openers^17,18; however, the mechanism is not clear. Vasodilations to both aprikalim and diazoxide were attenuated by glibenclamide, but dilations to SNP, papaverine, BK,⁸ and shear stress¹⁰ were not affected. Patch-clamp analysis shows that glibenclamide blocks K_ATP current in response to K_ATP openers in VSMCs from human coronary arteries.¹⁹ These findings indicate that K_ATP openers dilate HCAs by selectively opening K_ATP. We demonstrated that aprikalim-induced vasodilation was not affected by inhibiting NO synthase or cyclooxygenase using doses of antagonists that are effective in human vessels.^{8,10,11,20,21} This is similar to findings in rat cerebral and porcine coronary arterioles.^2,4 However, it contrasts with observations made in small cerebral arteries of diabetic rats.⁵

Hypoxia-Induced Vasodilation
Several animal studies have reported that K_ATP opening mediates hypoxia-induced vasodilation in conduit and resistance arteries.^1,2 However, inhibition of K_ATP is reported to be ineffective in attenuating hypoxia-induced vasorelaxation in rabbit aorta and porcine coronary arteries.^22,23 In the present study, glibenclamide significantly reduced hypoxic vasodilation in HCAs, suggesting that K_ATP are involved in transducing vascular signals to reduced tissue oxygen concentrations in the human heart.

In the present study, NO did not contribute to hypoxia-induced vasodilation. However, results from animal studies are varied, some indicating an important role for NO in isolated guinea pig hearts and rabbit coronary arteries^22,24 and others showing no effect in porcine cerebral arterioles and small coronary arteries.^2,25 The contribution of NO varies depending on the vascular bed, species, vessel size (conduit or resistance arteries), or experimental preparations. This highlights the importance of examining responses in vessels from human subjects.

An enhanced dilation to hypoxia at 5 minutes was observed in the presence of indomethacin. A similar augmentation has been described in porcine, monkey, and human coronary arteries.^20,26 Human endothelial cells showed the release of vasoconstrictor prostaglandins during the early phase of hypoxia.²¹

Hypoxia-induced vasodilation is endothelium-dependent in porcine small coronary arteries (170 µm diameter)², whereas it is endothelium-independent in porcine coronary arteries (1 mm diameter).²⁷ In human conduit coronary arteries, Siegel et al²⁸ reported that the contribution of endothelium to hypoxia-induced vasodilation is 49%, whereas Toda et al²⁰ reported no alteration after endothelial denudation. Our study identifies a unique response in the coronary microcirculation. We observed a direct hyperpolarizing and vasorelaxing effect of hypoxia on coronary smooth muscle cells that is endothelium-independent.

K_ATP Expression in the Coronary Microcirculation
The molecular profile of vascular K_ATP, especially in the coronary microcirculation, is not fully determined. Part of the difficulty arises from the complex nature of the channel, in that K_ATP are octamers composed of four Kir6.0 subfamily subunits (Kir6.1 and 6.2) and four SUR subunits (SUR1 and SUR2), combined into a heteromultimeric complex. K_ATP in VSMCs are proposed to comprise SUR2B and Kir6.1, based on similar electrophysiological characteristics of the reconstituted channels to the native one.²⁹ A recent study using a genetic mouse model lacking Kir6.1 showed the absence of K_ATP activity in VSMCs, suggesting a critical role of Kir6.1 for vascular K_ATP activity.³⁰ It is also reported that microvascular endothelial cells in the guinea pig heart express K_ATP composed of SUR2B and Kir6.1 and/or Kir6.2 subunits.³¹ These observations are consistent with our results showing the mRNA expression of Kir6.1, Kir6.2, and SUR2B and the localization of Kir6.1 protein in HCAs. RT-PCR revealed the expression of SUR1 and SUR2A in only one of three vessels tested. This may be attributed to presence of inflammatory cells or perivascular cells, including myocytes and neurons, which express SUR1 and SUR2A,^31,32 or a compensatory upregulation of this channel subunit. Additional studies should examine these possibilities.

DM and Impairment of K_ATP-Mediated Vasodilation
The mechanism of impaired vasodilation to K_ATP activation is not clear. We found impaired dilation to aprikalim and hypoxia but not to SNP and BK in HCAs from subjects with diabetes. Thus, the impairment seems to be specific for K_ATP mechanisms, because VSMC relaxation followed by decrease in intracellular Ca²⁺ concentration either by cGMP production attributable to NO (SNP) or by membrane hyperpolarization through Ca²⁺-activated K⁺ channel activation, as occurs with BK,⁸ was not reduced. As with other studies,^3–5 we observed reduced vasodilation to K_ATP opening. In contrast to our observations, an enhanced vasodilation to aprikalim was seen in diabetic canine coronary arterioles.⁶ This discrepancy might be dependent on the disease duration (months to years in rat models and humans versus hours to days in canine models) and/or the different experimental methods (in vitro versus in vivo).

Potential Problems
Sulfonylureas are clinically prescribed K_ATP blockers. In clinically relevant concentrations, glibenclamide increases basal coronary resistance^33,34 and attenuates vasodilation to hypoxia in the heart.^1,2 Because of the nature of the IRB approval, we are unable to retrospectively identify medication use. Some diabetic patients may have been treated with sulfonylureas, and it is possible that effect of these medicines could be responsible for the reduced dilation to aprikalim and hypoxia. For two reasons, we believe that such an effect, if present, was small and did not influence our interpretation of the results. First, the effect of aprikalim is fully reversible, because rinsing vessels with Krebs solution eliminated the inhibitory effect of glibenclamide on aprikalim-induced vasodilation. Multiple rinses during dissection and before experimentation would be expected to remove previously administered sulfonylureas. Second, the impaired dilation to aprikalim or hypoxia was seen in both type 1 and type 2 DM. It is unlikely that type 1 diabetic patients were treated with sulfonylureas. These reasons make it unlikely that retained or chronic effects of diabetic medications contributed to the impaired dilation to hypoxia and aprikalim.

In the present study, glibenclamide did not completely abolish vasodilation to hypoxia. Other factors may be involved in hypoxic dilation, as reported in other tissues, including Ca²⁺-activated K⁺ channels and cytochrome P450 metabolites,^25,35 decrease in tissue pH,³⁶ adenosine receptor activation,³⁷ or production of lactate.³⁸ An insufficient dose of glibenclamide is not likely, because the same dose markedly inhibited dilation to aprikalim. It is unlikely that non-K_ATP mechanisms are responsible for the reduced response to hypoxia, because the residual dilation to hypoxia after glibenclamide is similar in both DM and non-DM subjects.

Clinical Implications
DM is associated with an increased risk of cardiovascular and cerebrovascular morbidity and mortality. In addition to the more accelerated conduit coronary atherosclerosis, DM is associated with diffuse microvascular disease. Even for the same degree of atherosclerosis, patients with DM suffer greater morbidity from ischemia.³⁹ Clinical studies indicate that DM impairs myocardial perfusion.^40,41 Our results demonstrate impaired dilation to K_ATP opening in DM. This could impede vasodilator responses, especially during ischemia or hypoxia in patients with DM, possibly contributing to increased myocardial ischemic injury in these patients.

In summary, in HCAs, hypoxia-induced vasodilation is mediated largely by activation of K_ATP. Vasodilations to hypoxia and direct K_ATP opening are impaired in DM. This coronary microcirculatory dysfunction may contribute to the higher cardiovascular mortality and morbidity in DM.

	Acknowledgments

 
This study was supported by a Fellowship Grant from the American Heart Association (AHA), Iowa Affiliate, a National AHA Grant-in-Aid, a VA Merit Review award, and a grant from the NIH (HL52869).

Received July 10, 2002; revision received December 6, 2002; accepted December 6, 2002.

	References

Top Abstract Introduction Materials and Methods Results Discussion References

 

1. Daut J, Maier-Rudolph W, von Beckerath N, Mehrke G, Gunther K, Goedel-Meinen L. Hypoxic dilation of coronary arteries is mediated by ATP-sensitive potassium channels. Science. 1990; 247: 1341–1344.[Abstract/Free Full Text]
2. Liu Q, Flavahan NA. Hypoxic dilatation of porcine small coronary arteries: role of endothelium and KATP-channels. Br J Pharmacol. 1997; 120: 728–734.[CrossRef][Medline] [Order article via Infotrieve]
3. Cameron NE, Cotter MA. Impaired contraction and relaxation in aorta from streptozotocin-diabetic rats: role of polyol pathway. Diabetologia. 1992; 35: 1011–1019.[CrossRef][Medline] [Order article via Infotrieve]
4. Mayhan WG, Faraci FM. Responses of cerebral arterioles in diabetic rats to activation of ATP-sensitive potassium channels. Am J Physiol. 1993; 265: H152–H157.[Medline] [Order article via Infotrieve]
5. Zimmermann PA, Knot HJ, Stevenson AS, Nelson MT. Increased myogenic tone and diminished responsiveness to ATP-sensitive K⁺ channel openers in cerebral arteries from diabetic rats. Circ Res. 1997; 81: 996–1004.[Abstract/Free Full Text]
6. Kersten JR, Brooks LA, Dellsperger KC. Impaired microvascular response to graded coronary occlusion in diabetic and hyperglycemic dogs. Am J Physiol. 1995; 268: H1667–H1674.[Medline] [Order article via Infotrieve]
7. Miller FJ Jr, Dellsperger KC, Gutterman DD. Pharmacologic activation of the human coronary microcirculation in vitro: endothelium-dependent dilation and differential responses to acetylcholine. Cardiovasc Res. 1998; 38: 744–750.[Abstract/Free Full Text]
8. Miura H, Liu Y, Gutterman DD. Human coronary arteriolar dilation to bradykinin depends on membrane hyperpolarization: contribution of nitric oxide and Ca²⁺-activated K⁺ channels. Circulation. 1999; 99: 3132–3138.[Abstract/Free Full Text]
9. Miura H, Gutterman DD. Human coronary arteriolar dilation to arachidonic acid depends on cytochrome P-450 monooxygenase and Ca²⁺-activated K⁺ channels. Circ Res. 1998; 83: 501–507.[Abstract/Free Full Text]
10. Miura H, Wachtel RE, Liu Y, Loberiza FR Jr, Saito T, Miura M, Gutterman DD. Flow-induced dilation of human coronary arterioles: important role of Ca²⁺-activated K⁺ channels. Circulation. 2001; 103: 1992–1998.[Abstract/Free Full Text]
11. Terata K, Miura H, Liu Y, Loberiza F, Gutterman DD. Human coronary arteriolar dilation to adrenomedullin: role of nitric oxide and K⁺ channels. Am J Physiol Heart Circ Physiol. 2000; 279: H2620–H2626.[Abstract/Free Full Text]
12. Angus JA, Cocks TM, McPherson GA, Broughton A. The acetylcholine paradox: a constrictor of human small coronary arteries even in the presence of endothelium. Clin Exp Pharmacol Physiol. 1991; 18: 33–36.[Medline] [Order article via Infotrieve]
13. Lombard JH, Smeda J, Madden JA, Harder DR. Effect of reduced oxygen availability upon myogenic depolarization and contraction of cat middle cerebral artery. Circ Res. 1986; 58: 565–569.[Abstract/Free Full Text]
14. Karrer EE, Lincoln JE, Hogenhout S, Bennett AB, Bostock RM, Martineau B, Lucas WJ, Gilchrist DG, Alexander D. In situ isolation of mRNA from individual plant cells: creation of cell-specific cDNA libraries. Proc Natl Acad Sci U S A. 1995; 92: 3814–3818.[Abstract/Free Full Text]
15. Miller TR, Taber RD, Molinari EJ, Whiteaker KL, Monteggia LM, Scott VE, Brioni JD, Sullivan JP, Gopalakrishnan M. Pharmacological and molecular characterization of ATP-sensitive K⁺ channels in the TE671 human medulloblastoma cell line. Eur J Pharmacol. 1999; 370: 179–185.[CrossRef][Medline] [Order article via Infotrieve]
16. Davis-Taber R, Choi W, Feng J, Hoogenboom L, McNally T, Kroeger P, Shieh CC, Simmer R, Brioni JD, Sullivan JP, Gopalakrishnan M, Scott VE. Molecular characterization of human SUR2-containing K_ATP channels. Gene. 2000; 256: 261–270.[CrossRef][Medline] [Order article via Infotrieve]
17. Gollasch M, Bychkov R, Ried C, Behrendt F, Scholze S, Luft FC, Haller H. Pinacidil relaxes porcine and human coronary arteries by activating ATP-dependent potassium channels in smooth muscle cells. J Pharmacol Exp Ther. 1995; 275: 681–692.[Abstract/Free Full Text]
18. Bijlstra PJ, Lutterman JA, Russel FG, Thien T, Smits P. Interaction of sulphonylurea derivatives with vascular ATP-sensitive potassium channels in humans. Diabetologia. 1996; 39: 1083–1090.[Medline] [Order article via Infotrieve]
19. Gollasch M, Ried C, Bychkov R, Luft FC, Haller H. K⁺ currents in human coronary artery vascular smooth muscle cells. Circ Res. 1996; 78: 676–688.[Abstract/Free Full Text]
20. Toda N, Matsumoto T, Yoshida K. Comparison of hypoxia-induced contraction in human, monkey, and dog coronary arteries. Am J Physiol. 1992; 262: H678–H683.[Medline] [Order article via Infotrieve]
21. Michiels C, Arnould T, Knott I, Dieu M, Remacle J. Stimulation of prostaglandin synthesis by human endothelial cells exposed to hypoxia. Am J Physiol. 1993; 264: C866–C874.[Medline] [Order article via Infotrieve]
22. Jiang C, Collins P. Inhibition of hypoxia-induced relaxation of rabbit isolated coronary arteries by N^G-monomethyl-L-arginine but not glibenclamide. Br J Pharmacol. 1994; 111: 711–716.[Medline] [Order article via Infotrieve]
23. Shimizu S, Bowman PS, Thorne G III, Paul RJ. Effects of hypoxia on isometric force, intracellular Ca²⁺, pH, and energetics in porcine coronary artery. Circ Res. 2000; 86: 862–870.[Abstract/Free Full Text]
24. von Beckerath N, Cyrys S, Dischner A, Daut J. Hypoxic vasodilatation in isolated, perfused guinea-pig heart: an analysis of the underlying mechanisms. J Physiol (Lond). 1991; 442: 297–319.[Abstract/Free Full Text]
25. Leffler CW, Smith JS, Edrington JL, Zuckerman SL, Parfenova H. Mechanisms of hypoxia-induced cerebrovascular dilation in the newborn pig. Am J Physiol. 1997; 272: H1323–H1332.[Medline] [Order article via Infotrieve]
26. Rubanyi G, Paul RJ. Two distinct effects of oxygen on vascular tone in isolated porcine coronary arteries. Circ Res. 1985; 56: 1–10.[Abstract/Free Full Text]
27. Mellemkjaer S, Nielsen-Kudsk JE. Glibenclamide inhibits hypoxic relaxation of isolated porcine coronary arteries under conditions of impaired glycolysis. Eur J Pharmacol. 1994; 270: 307–312.[Medline] [Order article via Infotrieve]
28. Siegel G, Ruckborn K, Schnalke F, Muller J. Endothelial dysfunction in human atherosclerotic coronary arteries. Eur Heart J. 1993; 14 (suppl I): 99–103.
29. Yamada M, Isomoto S, Matsumoto S, Kondo C, Shindo T, Horio Y, Kurachi Y. Sulphonylurea receptor 2B and Kir6.1 form a sulphonylurea-sensitive but ATP-insensitive K⁺ channel. J Physiol (Lond). 1997; 499(pt 3): 715–720.[Medline] [Order article via Infotrieve]
30. Miki T, Suzuki M, Shibasaki T, Uemura H, Sato T, Yamaguchi K, Koseki H, Iwanaga T, Nakaya H, Seino S. Mouse model of Prinzmetal angina by disruption of the inward rectifier Kir6.1. Nat Med. 2002; 8: 466–472.[CrossRef][Medline] [Order article via Infotrieve]
31. Schnitzler MM, Derst C, Daut J, Preisig-Muller R. ATP-sensitive potassium channels in capillaries isolated from guinea- pig heart. J Physiol (Lond). 2000; 525(pt 2): 307–317.[Medline] [Order article via Infotrieve]
32. Karschin A, Brockhaus J, Ballanyi K. KATP channel formation by the sulphonylurea receptors SUR1 with Kir6.2 subunits in rat dorsal vagal neurons in situ. J Physiol. 1998; 509: 339–346.[Abstract/Free Full Text]
33. Samaha FF, Heineman FW, Ince C, Fleming J, Balaban RS. ATP-sensitive potassium channel is essential to maintain basal coronary vascular tone in vivo. Am J Physiol. 1992; 262: C1220–C1227.[Medline] [Order article via Infotrieve]
34. Farouque HMO, Worthley SG, Meredith IT, Skyrme-Jones RA, Zhang MJ. Effect of ATP-sensitive potassium channel inhibition on resting coronary vascular responses in humans. Circ Res. 2002; 90: 231–236.[Abstract/Free Full Text]
35. Armstead WM. Role of activation of calcium-sensitive K⁺ channels in NO- and hypoxia- induced pial artery vasodilation. Am J Physiol. 1997; 272: H1785–H1790.[Medline] [Order article via Infotrieve]
36. Ishizaka H, Kuo L. Acidosis-induced coronary arteriolar dilation is mediated by ATP-sensitive potassium channels in vascular smooth muscle. Circ Res. 1996; 78: 50–57.[Abstract/Free Full Text]
37. Minamino T, Kitakaze M, Komamura K, Node K, Takeda H, Inoue M, Hori M, Kamada T. Activation of protein kinase C increases adenosine production in the hypoxic canine coronary artery through the extracellular pathway. Arterioscler Thromb Vasc Biol. 1995; 15: 2298–2304.[Abstract/Free Full Text]
38. Mori K, Nakaya Y, Sakamoto S, Hayabuchi Y, Matsuoka S, Kuroda Y. Lactate-induced vascular relaxation in porcine coronary arteries is mediated by Ca²⁺-activated K⁺ channels. J Mol Cell Cardiol. 1998; 30: 349–356.[CrossRef][Medline] [Order article via Infotrieve]
39. Schernthaner G. Cardiovascular mortality and morbidity in type-2 diabetes mellitus. Diabetes Res Clin Pract. 1996; 31 (suppl): S3–S13.[CrossRef][Medline] [Order article via Infotrieve]
40. Nitenberg A, Valensi P, Sachs R, Dali M, Aptecar E, Attali JR. Impairment of coronary vascular reserve and ACh-induced coronary vasodilation in diabetic patients with angiographically normal coronary arteries and normal left ventricular systolic function. Diabetes. 1993; 42: 1017–1025.[Abstract]
41. Nahser PJ Jr, Brown RE, Oskarsson H, Winniford MD, Rossen JD. Maximal coronary flow reserve and metabolic coronary vasodilation in patients with diabetes mellitus. Circulation. 1995; 91: 635–640.[Abstract/Free Full Text]

This article has been cited by other articles:

				H. Kinoshita, N. Matsuda, H. Kaba, N. Hatakeyama, T. Azma, K. Nakahata, Y. Kuroda, K. Tange, H. Iranami, and Y. Hatano Roles of Phosphatidylinositol 3-Kinase-Akt and NADPH Oxidase in Adenosine 5'-Triphosphate-Sensitive K+ Channel Function Impaired by High Glucose in the Human Artery Hypertension, September 1, 2008; 52(3): 507 - 513. [Abstract] [Full Text] [PDF]

				N. Cheung, J. J. Wang, S. L. Rogers, F. Brancati, R. Klein, A. R. Sharrett, T. Y. Wong, and for the ARIC (Atherosclerosis Risk In Communities) Diabetic retinopathy and risk of heart failure. J. Am. Coll. Cardiol., April 22, 2008; 51(16): 1573 - 1578. [Abstract] [Full Text] [PDF]

				T. Kawano, K. Tanaka, K. Mawatari, S. Oshita, A. Takahashi, and Y. Nakaya Hyperglycemia Impairs Isoflurane-Induced Adenosine Triphosphate-Sensitive Potassium Channel Activation in Vascular Smooth Muscle Cells Anesth. Analg., March 1, 2008; 106(3): 858 - 864. [Abstract] [Full Text] [PDF]

				N. Cheung, J. J. Wang, R. Klein, D. J. Couper, A. R. Sharrett, and T. Y. Wong Diabetic Retinopathy and the Risk of Coronary Heart Disease: The Atherosclerosis Risk in Communities Study Diabetes Care, July 1, 2007; 30(7): 1742 - 1746. [Abstract] [Full Text] [PDF]

				S. Ghatta, R. R. Tunstall, S. Kareem, M. Rahman, and S. T. O'Rourke Sirolimus Causes Relaxation of Human Vascular Smooth Muscle: A Novel Action of Sirolimus Mediated via ATP-Sensitive Potassium Channels J. Pharmacol. Exp. Ther., March 1, 2007; 320(3): 1204 - 1208. [Abstract] [Full Text] [PDF]

				S. Takeda, T. Komaru, K. Takahashi, K. Sato, H. Kanatsuka, Y. Kokusho, K. Shirato, and H. Shimokawa Beating myocardium counteracts myogenic tone of coronary microvessels: involvement of ATP-sensitive potassium channels Am J Physiol Heart Circ Physiol, December 1, 2006; 291(6): H3050 - H3057. [Abstract] [Full Text] [PDF]

				P. T. Nowicki IL-1beta alters hemodynamics in newborn intestine: role of endothelin Am J Physiol Gastrointest Liver Physiol, September 1, 2006; 291(3): G404 - G413. [Abstract] [Full Text] [PDF]

				T. Szerafin, N. Erdei, T. Fulop, E. T. Pasztor, I. Edes, A. Koller, and Z. Bagi Increased Cyclooxygenase-2 Expression and Prostaglandin-Mediated Dilation in Coronary Arterioles of Patients With Diabetes Mellitus Circ. Res., September 1, 2006; 99(5): e12 - 317. [Abstract] [Full Text] [PDF]

				H. Kinoshita, T. Azma, H. Iranami, K. Nakahata, Y. Kimoto, M. Dojo, O. Yuge, and Y. Hatano Synthetic Peroxisome Proliferator-Activated Receptor-{gamma} Agonists Restore Impaired Vasorelaxation via ATP-Sensitive K+ Channels by High Glucose J. Pharmacol. Exp. Ther., July 1, 2006; 318(1): 312 - 318. [Abstract] [Full Text] [PDF]

				F. M. Lynch, C. Austin, A. M. Heagerty, and A. S. Izzard Adenosine and hypoxic dilation of rat coronary small arteries: roles of the ATP-sensitive potassium channel, endothelium, and nitric oxide Am J Physiol Heart Circ Physiol, March 1, 2006; 290(3): H1145 - H1150. [Abstract] [Full Text] [PDF]

				B. T. Larsen, H. Miura, O. A. Hatoum, W. B. Campbell, B. D. Hammock, D. C. Zeldin, J. R. Falck, and D. D. Gutterman Epoxyeicosatrienoic and dihydroxyeicosatrienoic acids dilate human coronary arterioles via BKCa channels: implications for soluble epoxide hydrolase inhibition Am J Physiol Heart Circ Physiol, February 1, 2006; 290(2): H491 - H499. [Abstract] [Full Text] [PDF]

				M. Zhou, O. Tanaka, M. Sekiguchi, H.-j. He, Y. Yasuoka, H. Itoh, K. Kawahara, and H. Abe ATP-sensitive K+-channel Subunits on the Mitochondria and Endoplasmic Reticulum of Rat Cardiomyocytes J. Histochem. Cytochem., December 1, 2005; 53(12): 1491 - 1500. [Abstract] [Full Text] [PDF]

				W. S. Park, J. Han, N. Kim, J.-H. Ko, S. J. Kim, and Y. E Earm Activation of inward rectifier K+ channels by hypoxia in rabbit coronary arterial smooth muscle cells Am J Physiol Heart Circ Physiol, December 1, 2005; 289(6): H2461 - H2467. [Abstract] [Full Text] [PDF]