Endothelium-dependent nitric oxide and hyperpolarization-mediated venous relaxation pathways in rat inferior vena cava

Transcription

1 From the American Venous Forum BASIC RESEARCH STUDIES Endothelium-dependent nitric oxide and hyperpolarization-mediated venous relaxation pathways in rat inferior vena cava Joseph D. Raffetto, MD, a,b,c Peng Yu, MD, PhD, a Ossama M. Reslan, MD, a Yin Xia, MD, PhD, a and Raouf A. Khalil, MD, PhD, a,b Boston and West Roxbury, Mass Introduction: The vascular endothelium plays a major role in the control of arterial tone; however, its role in venous tissues is less clear. The purpose of this study was to determine the role of endothelium in the control of venous function and the relaxation pathways involved. Methods: Circular segments of inferior vena cava (IVC) from male Sprague Dawley rats were suspended between two wires and isometric contraction to phenylephrine (Phe; 10 5 M) and 96 mm KCl was measured. Acetylcholine (Ach; to 10 5 M) was added and the percentage of venous relaxation was measured. To determine the role of nitric oxide (NO) and prostacyclin (PGI 2 ), vein relaxation was measured in the presence of the nitric oxide synthase inhibitor N -nitro-larginine methyl ester (L-NAME; M) and the cyclooxygenase inhibitor indomethacin (10 5 M). To measure the role of hyperpolarization, vein relaxation was measured in the presence of K channel activator cromakalim (10 11 to 10 6 M), and the nonselective K channel blocker tetraethylammonium (TEA; 10 3 M). To test for the contribution of a specific K channel, the effects of K channel blockers: glibenclamide (adenosine triphosphate [ATP]-sensitive K ATP, 10 5 M), 4-aminopyridine (4-AP; voltage-dependent K v,10 3 M), apamin (small conductance Ca 2 -dependent SK Ca, 10 7 M), and iberiotoxin (large conductance Ca 2 -dependent BK Ca,10 8 M) on Ach-induced relaxation were tested. Results: Ach caused concentration-dependent relaxation of Phe contraction (maximum %). Removal of endothelium abolished Ach-induced relaxation. IVC treatment with L-NAME partially reduced Ach relaxation ( %). In IVC treated with L-NAME plus indomethacin, significant Ach-induced relaxation ( %) could still be observed, suggesting a role of endothelium-derived hyperpolarizing factor (EDHF). In IVC treated with L-NAME, indomethacin and TEA, Ach relaxation was abolished, supporting a role of EDHF. In veins stimulated with high KCl, Ach caused relaxation (maximum %) that was abolished in the presence of L-NAME and indomethacin suggesting that any Ach-induced EDHF is blocked in the presence of high KCl depolarizing solution, which does not favor outward movement of K ion and membrane hyperpolarization. Cromakalim, an activator of K ATP, caused significant IVC relaxation when applied alone or on top of maximal Ach-induced relaxation, suggesting that the Ach response may not involve K ATP. Ach-induced relaxation was not inhibited by glibenclamide, 4-AP, or apamin, suggesting little role of K ATP,K v or SK Ca, respectively. In contrast, iberiotoxin significantly inhibited Ach-induced relaxation, suggesting a role of BK Ca. Conclusions: Thus, endothelium-dependent venous relaxation plays a major role in the control of venous function. In addition to NO, an EDHF pathway involving BK Ca may play a role in endothelium-dependent venous relaxation. (J Vasc Surg 2012;55: ) From the Division of Vascular and Endovascular Surgery, Brigham and Women s Hospital, a Harvard Medical School, b Boston; and VA Boston Healthcare System, West-Roxbury. c The study was partly supported by grants to Raouf A. Khalil from the National Heart, Lung, and Blood Institute (HL-65998, HL-98724), and The Eunice Kennedy Shriver National Institute of Child Health and Human Development (HD-60702). Author conflict of interest: none. Presented at the Twenty-third Annual Meeting of the American Venous Forum, San Diego, Calif, February 23-26, Reprint requests: Raouf A. Khalil, MD, PhD, Harvard Medical School, Brigham and Women s Hospital, Division of Vascular Surgery, 75 Francis Street, Boston, MA ( raouf_khalil@hms.harvard.edu). The editors and reviewers of this article have no relevant financial relationships to disclose per the JVS policy that requires reviewers to decline review of any manuscript for which they may have a conflict of interest /$36.00 Copyright 2012 by the Society for Vascular Surgery. Published by Elsevier Inc. All rights reserved. doi: /j.jvs

2 JOURNAL OF VASCULAR SURGERY Volume 55, Number 6 Raffetto et al 1717 Clinical Relevance: Endothelial dysfunction plays a major role in the pathogenesis of arterial disease and could also affect the course of venous disease. Although the endothelium-derived mediators have been well characterized in the arterial wall, the mechanisms of venous dilation are poorly understood. The present study in rat inferior vena cava, demonstrates that the mechanisms of venous relaxation partly involve the nitric oxide (NO) pathway, as well as a significant portion involving the hyperpolarization pathway. Although activation of ATP-sensitive K channels (K ATP ) could cause venous relaxation, Ach-induced endothelium-dependent relaxation seems to involve the large conductance Ca 2 -dependent K channel (BK Ca ). The identification of the mechanisms of venous relaxation could be important in the management of venous disease. Pharmacologic activators of the NO pathway and K channels could be useful in reducing vein restenosis and graft failure. On the other hand, pharmacologic therapy using specific blockers of the NO pathway and K channels could be useful in the management of primary and recurrent varicose veins. The vascular endothelium plays an important role in regulating arterial tone, and endothelial dysfunction could lead to progressive arterial disease such as hypertension and atherosclerosis. 1-4 An increase in shear stress on the endothelium triggers the release of endothelium-derived substances, including nitric oxide (NO), prostacyclin (PGI 2 ), and endothelium-derived hyperpolarizing factor (EDHF) Endothelial NO diffuses into vascular smooth muscle cells (VSMCs) where it activates guanylate cyclase and the NOcyclic guanosine monophosphate (cgmp) relaxation pathway, whereas PGI 2 activates adenylate cyclase and the PGI 2 -cyclic adenosine monophosphate (camp) relaxation pathway In some arteries, particularly the small resistance arteries, blocking the NO-cGMP pathway with the NO synthase inhibitor N -nitro-l-arginine methyl ester (L-NAME) and the PGI 2 -camp pathway with indomethacin does not completely inhibit endothelium-dependent acetylcholine (Ach) relaxation, suggesting the release of EDHF. 8,12,14 EDHF could cause hyperpolarization of endothelial cells, which propagates to VSMCs through myoendothelial gap junctions, or could cause the release of diffusible factors from the endothelium, which act on VSMCs leading to activation of K channels, hyperpolarization, and vascular relaxation. 1,8,15,16 Several types of K channels have been identified in the vasculature, including large conductance Ca 2 -activated (BK Ca ), intermediate conductance Ca 2 -activated (IK Ca ), small conductance Ca 2 -activated (SK Ca ), adenosine triphosphate (ATP)- sensitive (K ATP ), voltage-gated (K V ), and inward rectifier (K IR ). 8,16-18 The relative contribution of NO, PGI 2, and EDHF to arterial relaxation varies depending on the artery size and specific arterial bed. For instance, NO is a major factor in the relaxation of large arteries such as the aorta On the other hand, EDHF plays a major role in relaxation of small resistance arteries such as the mesenteric microvessels. 22,23 Although the pathways of vascular relaxation have been well-characterized in the arterial system, the role of NOcGMP, PGI 2 -camp, and EDHF in venous relaxation is less defined Also, while the majority of K channels have been characterized in the arterial system, little is known regarding the K channels involved in venous hyperpolarization and relaxation. The purpose of this study was to determine the role of the endothelium in the control of venous function and the potential endothelium-dependent venous relaxation pathways involved. METHODS Animals and tissues. Male Sprague-Dawley rats (12 weeks of age, g in weight; Charles River Laboratories, Wilmington, Mass) were maintained on ad libitum standard rat chow and tap water in 12-hour/12-hour light/dark cycle. Rats were euthanized by inhalation of carbon dioxide (CO 2 ). The inferior vena cava (IVC) was rapidly excised, placed in oxygenated Krebs solution, and carefully dissected and cleaned of connective tissue under microscopic visualization. The IVC was portioned into 3-mm rings in preparation for isometric contraction experiments. From each rat IVC, four 3-mm-wide segments were obtained. All vein segments were obtained from the IVC below the renal veins. Extreme care was taken throughout the tissue isolation and dissection procedure in order to minimize injury to the endothelium and the vein wall. One vein segment from each individual rat was used to perform one experiment and obtain one set of data points. Cumulative data from different vein segments from different rats were used to calculate the average data for each arm of the study. All procedures followed the National Institutes of Health (NIH) guide for the Care of Laboratory Animal Welfare Act and the guidelines of the Animal Care and Use Committee at Harvard Medical School. Isometric contraction. Circular segments of IVC were suspended between two stainless-steel hooks; one hook was fixed at the bottom of a tissue bath and the other hook was connected to a Grass force displacement transducer (FT03; Astro-Med, West Warwick, RI). Vein segments were stretched under 0.5 g of resting tension and allowed to equilibrate for 45 minutes in a tissue bath filled with 50-mL Krebs solution continuously bubbled with 95% O 2 5% CO 2 at 37 C. We have previously constructed the relationship between basal tension and the contraction to 96 mm KCl in rat IVC and demonstrated that 0.5-g basal tension produced maximal KCl contraction. 27 The changes in isometric contraction/relaxation were recorded on a Grass polygraph (Model 7D; Astro-Med), as previously described. 5 Control IVC contraction in response to 96-mM KCl was first elicited. Once the KCl maximum contraction was reached and a plateau achieved (within minutes) the tissue was washed three times in Krebs, 10 minutes each. The control contraction to 96-mM KCl was repeated twice before further experimentation. To investigate the venous endothelial function, IVC segments were precontracted with phenylephrine (Phe;

3 1718 Raffetto et al JOURNAL OF VASCULAR SURGERY June M) then treated with increasing concentrations (10 10 to 10 5 M) of Ach, and the percentage of venous relaxation was measured. To test the role of the endothelium, Ach-induced relaxation was compared in intact and endothelium-denuded IVC. In these experiments, the endothelium was mechanically removed by gently scraping the intimal surface of the IVC segment five times around the tip of forceps. 5 To test the role of endothelium-dependent NOcGMP relaxation pathway, experiments were repeated in IVC segments treated with the NO synthase (NOS) inhibitor L-NAME ( M) and the guanylate cyclase inhibitor 1H- 1,2,4 oxadiazolo[4,2-a]quinoxaline-1-one (ODQ; 10 5 M). To inhibit cyclooxygenase (COX), the veins were pretreated for 10 minutes with indomethacin (10 5 M). To test the role of hyperpolarization and potential release of EDHF, the effects of Ach (10 5 M) were tested on vein contraction induced by membrane depolarization by 96-mM KCl. To test for the potential K channel involved in hyperpolarization, the effects of Ach on Phe-induced IVC contraction were compared with those of the K channel opener and K ATP activator cromakalim (10 11 to 10 6 M). To further test for the specific K channel involved, Ach-induced relaxation of Phe contraction was examined in veins pretreated with the following K channel blockers: nonselective blocker of Ca 2 -dependent K channel tetraethylammonium (TEA; 10 3 M), K ATP blocker glibenclamide (10 5 M), K V blocker 4-aminopyridine (4-AP; 10 3 M), SK Ca blocker apamin (10 7 M), and BK Ca blocker iberiotoxin (10 8 M). 3,5-8,16,28 The concentrations of inhibitors of NOS, COX, and K channels were determined on the basis of previous studies and published IC 50. Solutions, drugs, and chemicals. Normal Krebs solution contained in mm: NaCl 120, KCl 5.9, NaHCO 3 25, NaH 2 PO 4 1.2, dextrose 11.5 (Fisher Scientific, Fair Lawn, NJ), CaCl (BDH Laboratory Supplies, Poole, England), and MgCl (Sigma, St. Louis, Mo). Krebs solution was bubbled with 95% O 2 and 5% CO 2 for 30 minutes at an adjusted ph 7.4. The 96-mM KCl was prepared as normal Krebs but with equimolar substitution of NaCl with KCl. Stock solutions of Phe (10 1 M), Ach (10 1 M), L-NAME (10 1 M), 4-AP (10 1 M), apamin (10 3 M; Sigma), and iberiotoxin (10 5 M) (Calbiochem, La Jolla, Calif) were prepared in distilled water. TEA (Sigma) was prepared as 1-mM solution in Krebs. Stock solutions of 1H- 1,2,4 oxadiazolo[4,2-a]quinoxaline-1-one (ODQ; 10 1 ; Calbiochem), indomethacin (10 1 M), cromakalim (10 2 M), and glibenclamide (10 1 M; Sigma) were prepared in diemtheylsulfoxide (DMSO). The final concentration of DMSO in the experimental solution was 0.1%. All other chemicals were of reagent grade or better. Statistical analysis. Cumulative data were analyzed and presented as means SEM with the n value representing the number of experiments on different vein segments from four to 12 rats. Data were compared using t-test for unpaired data and differences were considered statistically significant if P.05. Fig 1. Acetylcholine (Ach)-induced relaxation in rat inferior vena cava (IVC). Endothelium (Endo) intact IVC segments were precontracted with phenylephrine (Phe; 10 5 M), increasing Ach concentrations were added, the vein relaxation was observed (A), and the percentage of relaxation of Phe contraction was measured (B). In other experiments, the effects of Ach were measured in endothelium-denuded IVC (B). Data represent means SEM (n 12-31). * Significant (P.05). RESULTS Acetylcholine-induced endothelium-dependent relaxation. In endothelium-intact IVC, Phe (10 5 M) caused significant contraction ( g/mg tissue weight) that was maintained at steady-state for approximately 20 minutes. Topical application of increasing concentrations of Ach (10 9 to 10 5 M) caused concentration-dependent relaxation of Phe contraction that reached a maximum of % at 10 5 M. In endothelium-denuded IVC, Ach relaxation was completely absent ( %; P.001), indicating that the Ach-induced IVC relaxation is endothelium-dependent (Fig 1).

4 JOURNAL OF VASCULAR SURGERY Volume 55, Number 6 Raffetto et al 1719 Fig 2. Role of nitrous oxide (NO)-cyclic guanosine monophosphate (cgmp) in acetylcholine (Ach)-induced relaxation of rat inferior vena cava (IVC). IVC segments were either nontreated or pretreated with N -nitro-l-arginine methyl ester (L-NAME; M) or oxadiazolo quinoxaline (ODQ; 10 5 M) for 15 minutes. IVC segments were then precontracted with phenylephrine (Phe; 10 5 M), increasing Ach concentrations were added, and the percentage of relaxation of Phe contraction was measured. Data represent means SEM (n 8-31). * Significant (P.05). Fig 3. Contribution of nitrous oxide (NO), prostacyclin (PGI 2 ), and endothelium-derived hyperpolarizing factor (EDHF) to acetylcholine (Ach) relaxation of rat inferior vena cava (IVC). IVC segments were either nontreated or pretreated with N -nitro-larginine methyl ester (L-NAME; M) plus indomethacin (INDO; 10 5 M) or L-NAME INDO tetraethylammonium (TEA; 10 3 M) for 15 minutes. IVC segments were then precontracted with phenylephrine (Phe;10 5 M), increasing Ach concentrations were added, and the percentage of relaxation of Phe contraction was measured. The L-NAME-sensitive component of Ach relaxation ( 35%) is attributed to NO, whereas the TEAsensitive component ( 65%) is attributed to EDHF. Data represent means SEM (n 16-31). * Significant (P.05). Role of nitrous oxide-cyclic guanosine monophosphate. Pretreatment of IVC segments with the NOS blocker L- NAME ( M) caused a significant decrease in Ach-induced relaxation to a maximum of % (P.01) at 10 5 M. Also, pretreatment of IVC segments with the guanylate cyclase inhibitor ODQ (10 5 M) caused significant reduction of Ach relaxation to a maximum of % (P.001) at 10 5 M(Fig 2). These data indicated a role for NO-cGMP pathway in IVC relaxation. However, Ach-induced relaxation was not completely inhibited by either L-NAME or ODQ, suggesting the involvement of other venous relaxation pathways. Contribution of prostacyclin and endotheliumderived hyperpolarizing factor. To evaluate the role of PGI 2 in the Ach-induced relaxation observed during NO blockade by L-NAME, veins were treated with the COX inhibitor indomethacin (INDO; 10 5 M). Compared to control Ach-induced relaxation ( %), in IVC treated with both L-NAME INDO, Ach caused significantly less relaxation that reached a maximum of % at 10 5 M(Fig 3). Ach-induced relaxation in the presence of L-NAME INDO was not significantly different from that in the presence of L-NAME alone, suggesting little role of PGI 2. In IVC treated with L-NAME INDO nonselective Ca 2 dependent K channel blocker TEA, Ach-induced relaxation was abolished ( %), suggesting a role of EDHF (Fig 3). Effect of acetylcholine on KCl-induced contraction. High 96-mM KCl depolarizing solution creates a K gradient that does not favor outward movement of K ion and hyperpolarization. 25,29 KCl caused significant IVC contraction ( g/mg tissue weight) that was maintained at steady-state for approximately 20 minutes. Ach caused concentration-dependent relaxation of KClinduced contraction that reached a maximum of % at 10 5 M. In IVC treated with L-NAME INDO, Ach-induced relaxation of KCl contraction was significantly reduced ( %; P.0001; Fig 4). The absence of Ach-induced relaxation of KCl contraction in IVC treated with L-NAME INDO suggests that any Ach-induced EDHF and hyperpolarization is blocked in the presence of high KCl depolarizing solution. Effect of K ATP channel opener cromakalim. The prototype K ATP channel activator cromakalim caused concentration-dependent relaxation of Phe contraction that reached a maximum of % at 10 6 M(Fig 5, A). In contrast, with Ach-induced relaxation of KCl contraction, cromakalim caused minimal relaxation of KClinduced relaxation ( % at 10 6 M; Fig 5, B), supporting that hyperpolarization is an important relaxation mechanism in IVC segments. Interaction between nitrous oxide-dependent and hyperpolarization-dependent vein relaxation. In IVC treated with L-NAME INDO, Ach still caused significant relaxation of Phe contraction, and addition of cromakalim (10 6 M) caused further relaxation of Phe contraction to % (P.001; Fig 6). Because cromakalim activates K ATP, its additive effects to maximal

5 1720 Raffetto et al JOURNAL OF VASCULAR SURGERY June 2012 Fig 4. Contribution of nitrous oxide (NO), prostacyclin (PGI 2 ), and endothelium-derived hyperpolarizing factor (EDHF) to acetylcholine (Ach) relaxation of depolarization-induced contraction of rat inferior vena cava (IVC). IVC segments were either nontreated or pretreated with N -nitro-l-arginine methyl ester (L- NAME; M) plus indomethacin (INDO;10 5 M) for 15 minutes. IVC segments were then precontracted with high KCl (96 mm) depolarizing solution, increasing Ach concentrations were added, and the percentage of relaxation of KCl contraction was measured. The L-NAME-sensitive component of Ach relaxation of KCl contraction ( 95%) is attributed to NO, and almost no relaxation is attributed to EDHF. Data represent means SEM (n 12). * Significant (P.05). Ach-induced relaxation suggest that Ach may activate K channels other than K ATP. In comparison, in IVC treated with L-NAME INDO, cromakalim caused relaxation of Phe contraction ( %), but was unexpectedly less than the relaxation caused by cromakalim alone ( %), suggesting cross-talk between the NO-cGMP and hyperpolarization pathways. On the other hand, Ach did not cause any further significant relaxation of Phe contraction in veins treated with cromakalim in the presence of L-NAME INDO (Fig 6), likely due to cromakalim-induced maximal hyperpolarization. Effect of selective K channel blockers. Ach could use one or more K channel to produce IVC hyperpolarization. 8,16-18 In order to determine the K channel(s) involved in the IVC relaxation, selective K channel blockers were used. Compared to control Ach-induced relaxation, Ach-induced relaxation was not significantly reduced in IVC treated with K ATP blocker glibenclamide, K v blocker 4-AP, or SK Ca blocker apamin. In contrast, Achinduced relaxation was significantly reduced in IVCs treated with the BK Ca blocker IbTx (Fig 7). DISCUSSION The present findings in rat IVC demonstrate that: (1) Ach causes significant endothelium-dependent venous relaxation; (2) approximately one-third of Ach-induced venous relaxation involves NO synthesis and the NO-cGMP pathway; (3) hyperpolarization is an important mechanism of venous relaxation; (4) approximately two-thirds of Achinduced venous relaxation involves activation of BK Ca - dependent hyperpolarization pathway; and (5) there is a cross-talk between the hyperpolarization-dependent and NO-cGMP pathways during vein relaxation. Several groups have examined endothelium-dependent relaxation in numerous arterial preparations However, our knowledge regarding the endothelium-dependent venous relaxation pathways is limited This is in part because the venous tissue is very delicate and difficult to handle. We have previously shown that the rat IVC produces significant and measurable contraction. 5,34 Consistent with our previous reports, we found that the rat IVC produced significant Phe contraction. The Phe contraction was maintained for at least 20 minutes, making it possible to measure Ach-induced vein relaxation. Consistent with previous reports in arterial segments, 35,36 Ach-induced IVC relaxation was concentration dependent and reached a maximum at M concentration. Also, the magnitude of Ach-induced venous relaxation (50% to 60%) was in the range of Ach-induced relaxation in arterial segments. Ach-induced venous relaxation was endothelium dependent because it was nearly abolished in endotheliumdenuded veins. Our results are consistent with an earlier study demonstrating endothelium-dependent relaxation in rat femoral veins. 25 These observations suggest that the venous endothelium is functional and produces significant relaxation of venous tissue. The endothelium is known to release several vasodilator substances, including NO, PGI 2, and EDHF. 5-8 Endothelial-derived vasodilators are released in response to shear stress acting on endothelial cells, and also in response to neurohumoral mediators in the circulation acting on specific receptors on the endothelium, including growth factors, cathecholamines, histamine, Ach, arachidonic acid, bradykinin, serotonin, thrombin, endothelin, arginine vasopressin, and adenosine diphosphate released by aggregating platelets. 8 In large blood vessels such as the aorta, pharmacologic inhibitors of NOS and COX almost abolish Ach-induced relaxation. In contrast, in small resistance arteries pretreated with NOS and COX inhibitors, a significant component of Ach-induced relaxation is still observed, suggesting a role of EDHF in the regulation of arterial tone. 1,8 Endothelium-derived NO is known to diffuse into vascular smooth muscle (VSM) where it activates guanylate cyclase and increases cgmp production. cgmp activates cgmp-dependent protein kinase (PKG), which promotes Ca 2 extrusion and decreases VSM [Ca 2 ] i and also causes phosphorylation and inactivation of myosin light chain kinase leading to inhibition of VSM contraction and consequent vascular relaxation. 11,13 We have previously shown that the rat IVC produces significant amounts of NO under basal conditions and that Ach produces significant increases in NO production. 5 Consistent with our previous report, we found that Ach-induced IVC relaxation was partly

6 JOURNAL OF VASCULAR SURGERY Volume 55, Number 6 Raffetto et al 1721 Fig 5. Contribution of hyperpolarization pathway to relaxation of rat inferior vena cava (IVC). IVC segments were precontracted with either phenylephrine (Phe; 10 5 M) (A) or high KCl (96 mm) depolarizing solution (B). IVC segments were then treated with increasing concentrations of the adenosine triphosphate-sensitive K channels (K ATP ) channel activator cromakalim, or acetylcholine (Ach) for comparison. Compared to the Ach response, cromakalim caused larger relaxation of Phe-induced contraction, but almost no relaxation of depolarization-induced KCl contraction. Data represent means SEM (n 8-31). Fig 6. Interaction between nitrous oxide (NO)-dependent and hyperpolarization-dependent relaxation in rat inferior vena cava (IVC). IVC segments were precontracted with phenylephrine (Phe; 10 5 M) and a control acetylcholine (Ach;10 5 M) or cromakalim (10 6 M) relaxation was elicited (open bars). In parallel experiments, IVC segments were pretreated with N -nitro-l-arginine methyl ester (L-NAME; M) indomethacin (INDO; 10 5 M) for 15 minutes, precontracted with Phe, then treated with Ach (10 5 M) or cromakalim (10 6 M; dotted bars). At steady-state, the veins treated with Ach were further challenged with cromakalim, whereas the veins treated with cromakalim were further challenged with Ach, and the veins were observed for any further relaxation (solid bars). Cromakalim caused further relaxation of Ach-treated veins, while Ach did not cause any further significant relaxation of cromakalim-treated veins. Data represent means SEM (n 7-31). *Measurements in L-NAME INDO treated veins are significantly different (P.05) from corresponding measurements in control veins. # Measurements in veins treated with L-NAME INDO Ach followed by cromakalim are significantly different (P.05) from corresponding measurements in control veins. blocked by the NOS inhibitor L-NAME. Also, inhibition of guanylate cyclase by ODQ caused significant decrease in Ach-induced relaxation, indicating the importance of NOcGMP pathway in venous relaxation. 1,37,38 However, a significant Ach-induced relaxation could still be observed in L-NAME-treated IVC, suggesting the involvement of other endothelium-derived relaxing factors (EDRFs). Pretreatment of IVC with L-NAME plus the COX inhibitor INDO did not cause any further reduction in Ach-induced relaxation, suggesting little role of PGI 2. This study in rat IVC demonstrated that one-third of Ach relaxation was due to NO, and by excluding PGI 2, the remaining two-thirds of Ach-induced relaxation seem to involve EDHF. One way to block EDHF is to use high KCl depolarizing solution, which creates a K gradient that does not favor outward movement of the K ion and, therefore, prevents hyperpolarization. 25,29 Ach caused approximately 60% relaxation of KCl contraction, that was abolished in IVC treated with L-NAME INDO, further supporting a role of the NOcGMP pathways. Importantly, in IVC treated with both L-NAME and INDO, Ach still caused relaxation of Phe contraction, but failed to cause relaxation of KCl contraction. These data support a role of EDHF in Ach-induced relaxation and that this hyperpolarization factor is blocked in the presence of KCl depolarizing solution. These data also support that EDHF is an important pathway, in not only the regulation of arterial tone, 25,26 but also the regulation of venous function. Although the exact nature of the EDHF molecules is not exactly known, several candidate compounds have been studied and likely function as EDRF and lead to hyperpolarization. Potassium ions are known to accumulate in the intercellular space between endothelial and smooth muscle cells (SMCs). The potassium ions form what is called a potassium cloud that activates SK Ca and IK Ca on the endothelium and hyperpolarization ensues by activating both K IR and Na /K pump on the SMC. 10 Other important EDHFs are metabolites of arachidonic acid produced by cytochrome P450 monooxygenase and are called epoxyei-

7 1722 Raffetto et al JOURNAL OF VASCULAR SURGERY June 2012 Fig 7. Effect of selective K channel blockers on acetylcholine (Ach) relaxation of rat inferior vena cava (IVC). IVC segments were precontracted with phenylephrine (Phe; 10 5 M) and a control Ach (10 5 M)-induced relaxation was elicited. At steadystate, the veins were treated with a specific K channel blocker: adenosine triphosphate-sensitive K channels (K ATP ) blocker glibenclamide (10 5 M), K V blocker 4-aminopyridine (4-AP; 10 3 M), SK Ca blocker apamin (10 7 M), and BK Ca blocker iberiotoxin (IbTX;10 8 M), and the remaining Ach relaxation was measured. Data represent means SEM (n 4-31). *Measurements in veins treated with K channel blocker are significantly different (P.05) from corresponding measurements in control veins. cosatrienoic (EET) acid. 39 EETs are known to hyperpolarize smooth muscle by activating BK Ca and other K channels. 40 Other EDHF candidates are hydrogen peroxide that may activate K channels on the endothelium or SMC, 15 but the mechanism is unclear. Additionally, C-type natriuretic peptide leads to hyperpolarization of both arterial and venous tissue by activating BK Ca. 8,26 The current study was not intended to characterize the EDHF molecule, but to highlight the role of hyperpolarization and potential K channel(s) involved in IVC relaxation. There are several K channel openers that include cromakalim, pinacidil, nicorandil, diazoxide, and RP Cromakalim works mainly by activating K ATP channels. 41 Our data demonstrate significant IVC relaxation in response to cromakalim, supporting a role of hyperpolarization and activation of K ATP channels as a major mechanism of IVC relaxation. This is consistent with other studies demonstrating a significant relaxation to cromakalim in human saphenous vein. 42,43 Importantly, the venorelaxant effect of cromakalim was essentially abolished in IVC precontracted with 96-mM KCl. High KCl solution abrogates the outward flow of K from the endothelium and SMC via K channels, and, hence, completely eliminates the possibility for hyperpolarization and relaxation. Our findings are consistent with previous reports in both rat and human venous tissues. 25,42 In search for the potential K channel involved in IVC relaxation, we found that Ach-induced relaxation was not reversed in the presence of specific blockers of K ATP,K V,or SK Ca. In contrast, Ach-induced IVC relaxation was abolished in the presence of the BK Ca blocker iberiotoxin. These findings suggest that a key modulator of hyperpolarization in the IVC venous system is via the activation of BK Ca channel. Our data are consistent with reports that human umbilical vein endothelial cells express BK Ca. 44 Thus, although in the arterial system, EDHF may require the activation of SK Ca and also IK Ca to cause hyperpolarization, 3,8,45,46 this may not be the situation in veins, particularly the IVC. NO not only activates guanylate cyclase to produce cgmp, but can also coactivate various K channels, including K ATP,K V, and BK Ca. 1,8 In porcine internal mammary and rat mesenteric arteries, Ach-induced relaxation via the NO-cGMP pathway could also involve activation of SK Ca, IK Ca, and K V channels. 37,47 We found that cromakalim caused less relaxation in the presence of L-NAME and INDO than cromakalim alone. This would indicate that NO, either directly or indirectly via increasing cgmp, is an important effector molecule in enhancing K ATP channel activity during hyperpolarization and relaxation of rat IVC (Fig 8). It is important to note that the present study was conducted on rat IVC. Although the rat IVC is a very thin and delicate preparation, we have perfected the vein harvesting and dissection technique and obtained reproducible data and consistent findings. 5,27,48,49 Our tissue histology studies demonstrate that, similar to human veins, the rat IVC has three anatomical layers; intima, media, and adventitia. 34 Also, physiologic vein function studies demonstrate that the rat IVC contract and relax to various constrictor and vasodilator stimuli. 5,49 Although the rat is a four-legged animal, studies on rat tissues avoid variability in age, body weight, and other confounding factors in humans. To enhance the relevance of the present findings to lower extremity veins, studies should be conducted on rat iliac and femoral vein. However, because human lower extremity veins are subjected to more pressure in the upright posture and may adapt differently to pressure/stretch, one needs to be careful in extrapolating the present data in rat veins to human veins. We have previously studied vein functions in circular segments of human saphenous veins and varicose veins. 50 We have shown that membrane depolarization by high 96-mM KCl solution causes significant contraction of human saphenous veins. Whether membrane hyperpolarization plays a role in human vein relaxation is unclear and should represent an important area for future investigations. Furthermore, to enhance the translational aspects, future studies should characterize not only the relaxation pathways in control greater saphenous veins but also the changes in the venous relaxation mechanisms in varicose veins. Venous wall disease could take several forms ranging from vein restenosis and graft failure 51 to venous dilation and varicose veins. 52 Dilated varicose veins, both primary and recurrent, are known to involve valve dysfunction with increased venous pressure leading to wall dilation and refluxing valves in any segment of the superficial and deep system. 53 However, venous wall dilation could also precede valve dysfunction and may have an equally important role in the pathophysiology of varicose vein development. 52,54 The identification of the mechanisms underlying venous relaxation could be important in un-

8 JOURNAL OF VASCULAR SURGERY Volume 55, Number 6 Raffetto et al 1723 Fig 8. Venous relaxation pathways in rat inferior vena cava (IVC). Acetylcholine (Ach) stimulates the nitrous oxide (NO)-cyclic guanosine monophosphate (cgmp) pathway to cause vein relaxation. Ach also activates BK Ca leading to hyperpolarization and further vein relaxation. Cromakalim activates adenosine triphosphate (ATP)-sensitive K channels (K ATP ) causing hyperpolarization and vein relaxation. NO-cGMP could directly or indirectly enhance K ATP activity via a cross-talk mechanism. BK Ca, Large conductance Ca 2 -activated K channel; EDHF, endothelium-derived hyperpolarizing factor; enos, endothelial nitric oxide synthase; VSMC, vascular smooth muscle cell. derstanding the pathophysiologic mechanisms underlying the vein wall stenosis associated with graft failure and vein wall dilation associated with varicose vein formation. Targeting these abnormal venodilator mechanisms could be useful in the management of venous disease. Pharmacologic activators of the NO pathway and K channels could be useful in reducing vein restenosis and graft failure. On the other hand, pharmacologic therapy using specific inhibitors of the NO pathway and specific K channel blockers could be useful in the management of primary and recurrent varicose veins. Vein valvular disease has been suggested to cause vein reflux and subsequent venous dilation. However, venous wall duplex ultrasonographic studies have described an ascending phenomenon to the pathophysiology of varicose veins, suggesting that vein wall dilation may precede valvular dysfunction. 52 Also, biochemical abnormalities similar to those observed in varicose veins have been demonstrated in normal-appearing human saphenous vein wall adjacent to varicose veins and between competent venous valves. 55,56 We have also shown that prolonged increases in vein wall tension in rat IVC is associated with venous dysfunction and decreased vein contraction. 27 The increased vein wall tension causes an increase in matrix metalloproteinases (MMP)-2 and MMP-9, and MMPs in turn cause vein wall hyperpolarization and venous relaxation possibly by activating large conductance Ca 2 -dependent K channels. 5 MMP-2-induced hyperpolarization could cause inhibition of Ca 2 entry into venous smooth muscle. 34 The present study supports the contention that hyperpolarization is a major relaxation pathway in veins. We postulate that persistent and recalcitrant venous hypertension leads to increased vein wall tension and release of MMP-2 and MMP-9, which in turn cause hyperpolarization and inhibition of Ca 2 entry into venous smooth muscle and leads to decreased contraction and the increased venous dilation associated with varicose veins. 27 It is likely that both valve dysfunction and vein wall dilation play a role in the pathophysiology of varicose veins. Thus, while removing reflux would lead to remodeling and shrink down of varicose veins in humans, it would also remove the persistent venous hypertension and the increased vein wall tension. This will in turn reverse the increase in MMPs, and the persistent vein wall hyperpolarization and the venous dilation associated with varicose veins. In conclusion, the present study elucidated some of the mechanisms involved in relaxation of rat IVC. Importantly, venous relaxation involves endothelium-dependent pathways that are mediated by both NO-cGMP and EDHF, with an important contribution of hyperpolarization via BK Ca and K ATP channels. Further research will be required to determine the role of NO-cGMP, PGI 2 -camp, and EDHF relaxation pathways in human saphenous veins and varicose veins and to examine potential pharmacologic interventions that can be useful in the treatment of the overdistended varicose veins.

9 1724 Raffetto et al JOURNAL OF VASCULAR SURGERY June 2012 AUTHOR CONTRIBUTIONS Conception and design: JR, RAK Analysis and interpretation: JR, RK Data collection: JR, PY, OR, YX Writing the article: JR, RK Critical revision of the article: JR, RK Final approval of the article: JR, RK Statistical analysis: JR, RK Obtained funding: RK Overall responsibility: RK REFERENCES 1. Félétou M, Köhler R, Vanhoutte PM. Endothelium-derived vasoactive factors and hypertension: possible roles in pathogenesis and as treatment targets. Curr Hypertens Rep 2010;12: Vanhoutte PM, Shimokawa H, Tang EH, Feletou M. Endothelial dysfunction and vascular disease. Acta Physiol (Oxf) 2009;196: Grgic I, Kaistha BP, Hoyer J, Köhler R. Endothelial Ca -activated K channels in normal and impaired EDHF-dilator responses relevance to cardiovascular pathologies and drug discovery. Br J Pharmacol 2009; 157: Vanhoutte PM. Endothelial dysfunction: the first step toward coronary arteriosclerosis. Circ J 2009;73: Raffetto JD, Ross RL, Khalil RA. Matrix metalloproteinase 2-induced venous dilation via hyperpolarization and activation of K channels: relevance to varicose vein formation. J Vasc Surg 2007;45: Ross GR, Yallampalli C. Endothelium-independent relaxation by adrenomedullin in pregnant rat mesenteric artery: role of camp-dependent protein kinase A and calcium-activated potassium channels. J Pharmacol Exp Ther 2006;317: Köhler R, Hoyer J. The endothelium-derived hyperpolarizing factor: insights from genetic animal models. Kidney Int 2007;72: Félétou M, Vanhoutte PM. Endothelium-derived hyperpolarizing factor: where are we now? Arterioscler Thromb Vasc Biol 2006;26: Brandes RP, Schmitz-Winnenthal FH, Félétou M, Gödecke A, Huang PL, Vanhoutte PM, et al. An endothelium-derived hyperpolarizing factor distinct from NO and prostacyclin is a major endotheliumdependent vasodilator in resistance vessels of wild-type and endothelial NO synthase knockout mice. Proc Natl Acad Sci USA2000;97: Edwards G, Dora KA, Gardener MJ, Garland CJ, Weston AH. K is an endothelium-derived hyperpolarizing factor in rat arteries. Nature 1998;396: Fleming I, Busse R. NO: the primary EDRF. J Mol Cell Cardiol 1999;31: Parkington HC, Coleman HA, Tare M. Prostacyclin and endotheliumdependent hyperpolarization. Pharmacol Res 2004;49: Ignarro LJ, Kadowitz PJ. The pharmacological and physiological role of cyclic GMP in vascular smooth muscle relaxation. Annu Rev Pharmacol Toxicol 1985;25: Busse R, Edwards G, Félétou M, Fleming I, Vanhoutte PM, Weston AH. EDHF: bringing the concepts together. Trends Pharmacol Sci 2002;23: Shimokawa H, Morikawa K. Hydrogen peroxide is an endotheliumderived hyperpolarizing factor in animals and humans. J Mol Cell Cardiol 2005;39: Edwards G, Félétou M, Weston AH. Endothelium-derived hyperpolarising factors and associated pathways: a synopsis. Pflugers Arch 2010; 459: Coleman HA, Tare M, Parkington HC. Endothelial potassium channels, endothelium-dependent hyperpolarization and the regulation of vascular tone in health and disease. Clin Exp Pharmacol Physiol 2004; 31: Zygmunt PM, Edwards G, Weston AH, Larsson B, Högestätt ED. Involvement of voltage-dependent potassium channels in the EDHFmediated relaxation of rat hepatic artery. Br J Pharmacol 1997;121: Al-Zobaidy MJ, Craig J, Martin W. Differential sensitivity of basal and acetylcholine-induced activity of nitric oxide to blockade by asymmetric dimethylarginine in the rat aorta. Br J Pharmacol 2010; 160: Zembowicz A, Hatchett RJ, Jakubowski AM, Gryglewski RJ. Involvement of nitric oxide in the endothelium-dependent relaxation induced by hydrogen peroxide in the rabbit aorta. Br J Pharmacol 1993;110: Davis JR, Giardina JB, Green GM, Alexander BT, Granger JP, Khalil RA. Reduced endothelial NO-cGMP vascular relaxation pathway during TNF-alpha-induced hypertension in pregnant rats. Am J Physiol Regul Integr Comp Physiol 2002;282:R Gerber RT, Anwar MA, Poston L. Enhanced acetylcholine induced relaxation in small mesenteric arteries from pregnant rats: an important role for endothelium-derived hyperpolarizing factor (EDHF). Br J Pharmacol 1998;125: Brøndum E, Kold-Petersen H, Simonsen U, Aalkjaer C. NS309 restores EDHF-type relaxation in mesenteric small arteries from type 2 diabetic ZDF rats. Br J Pharmacol 2010;159: Komori K, Shimokawa H, Vanhoutte PM. Endothelium-dependent relaxation in response to aggregating platelets in porcine femoral veins and its modulation by diet. Circulation 1989;80: Nagao T, Vanhoutte PM. Hyperpolarization contributes to endotheliumdependent relaxations to acetylcholine in femoral veins of rats. Am J Physiol 1991;261(4 Pt 2):H Banks M, Wei CM, Kim CH, Burnett JC Jr, Miller VM. Mechanism of relaxations to C-type natriuretic peptide in veins. Am J Physiol 1996; 271(5 Pt 2):H Raffetto JD, Qiao X, Koledova VV, Khalil RA. Prolonged increases in vein wall tension increase matrix metalloproteinases and decrease constriction in rat vena cava: potential implications in varicose veins. J Vasc Surg 2008;48: Kitamura K, Kamouchi M. K channel openers activate different K channels in vascular smooth muscle cells. Cardiovasc Drugs Ther 1993;7 Suppl 3: Weston AH, Longmore J, Newgreen DT, Edwards G, Bray KM, Duty S. The potassium channel openers: a new class of vasorelaxants. Blood Vessels 1990;27: Ng KF, Leung SW, Man RY, Vanhoutte PM. Endothelium-derived hyperpolarizing factor mediated relaxations in pig coronary arteries do not involve Gi/o proteins. Acta Pharmacol Sin 2008;29: Shi Y, Ku DD, Man RY, Vanhoutte PM. Augmented endotheliumderived hyperpolarizing factor-mediated relaxations attenuate endothelial dysfunction in femoral and mesenteric, but not in carotid arteries from type I diabetic rats. J Pharmacol Exp Ther 2006;318: Zellers TM, Wu YQ, McCormick J, Vanhoutte PM. Prostacyclininduced relaxations of small porcine pulmonary arteries are enhanced by the basal release of endothelium-derived nitric oxide through an effect on cyclic GMP-inhibited-cyclic AMP phosphodiesterase. Acta Pharmacol Sin 2000;21: Büssemaker E, Popp R, Binder J, Busse R, Fleming I. Characterization of the endothelium-derived hyperpolarizing factor (EDHF) response in the human interlobar artery. Kidney Int 2003;63: Raffetto JD, Barros YV, Wells AK, Khalil RA. MMP-2 induced vein relaxation via inhibition of [Ca2 ]e-dependent mechanisms of venous smooth muscle contraction. Role of RGD peptides. J Surg Res 2010; 159: Gauthier KM, Goldman DH, Aggarwal NT, Chawengsub Y, Falck JR, Campbell WB. Role of arachidonic acid lipoxygenase metabolites in acetylcholine-induced relaxations of mouse arteries. Am J Physiol Heart Circ Physiol 2011;300:H Lu X, Kassab GS. Assessment of endothelial function of large, medium, and small vessels: a unified myograph. Am J Physiol Heart Circ Physiol 2011;300:H94-H Favaloro JL, Kemp-Harper BK. Redox variants of NO (NO{middle dot} and HNO) elicit vasorelaxation of resistance arteries via distinct mechanisms. Am J Physiol Heart Circ Physiol 2009;296:H

10 JOURNAL OF VASCULAR SURGERY Volume 55, Number 6 Raffetto et al Pimentel AM, Costa CA, Carvalho LC, Brandão RM, Rangel BM, Tano T, et al. The role of NO-cGMP pathway and potassium channels on the relaxation induced by clonidine in the rat mesenteric arterial bed. Vascul Pharmacol 2007;46: Fleming I, Busse R. Endothelium-derived epoxyeicosatrienoic acids and vascular function. Hypertension 2006;47: McSherry IN, Sandow SL, Campbell WB, Falck JR, Hill MA, Dora KA. A role for heterocellular coupling and EETs in dilation of rat cremaster arteries. Microcirculation 2006;13: Glavind-Kristensen M, Matchkov V, Hansen VB, Forman A, Nilsson H, Aalkjaer C. KATP-channel-induced vasodilation is modulated by the Na,K-pump activity in rabbit coronary small arteries. Br J Pharmacol 2004;143: de Moura RS, Jasbik W. Actions of cromakalim in isolated human saphenous vein. Hypertension 1992;19(2 Suppl):II Beattie DK, Gosling M, Davies AH, Powell JT. The effects of potassium channel openers on saphenous vein exposed to arterial flow. Eur J Vasc Endovasc Surg 1998;15: Köhler R, Schönfelder G, Hopp H, Distler A, Hoyer J. Stretchactivated cation channel in human umbilical vein endothelium in normal pregnancy and in preeclampsia. J Hypertens 1998;16: Eichler I, Wibawa J, Grgic I, Knorr A, Brakemeier S, Pries AR, et al. Selective blockade of endothelial Ca2 -activated small- and intermediateconductance K -channels suppresses EDHF-mediated vasodilation. Br J Pharmacol 2003;138: Köhler R, Wulff H, Eichler I, Kneifel M, Neumann D, Knorr A, et al. Blockade of the intermediate-conductance calcium-activated potassium channel as a new therapeutic strategy for restenosis. Circulation 2003; 108: Pagán RM, Prieto D, Hernández M, Correa C, García-Sacristán A, Benedito S, et al. Regulation of NO-dependent acetylcholine relaxation by K channels and the Na -K ATPase pump in porcine internal mammary artery. Eur J Pharmacol 2010;641: Raffetto JD, Khalil RA. Ca(2 )-dependent contraction by the saponoside escin in rat vena cava: implications in venotonic treatment of varicose veins. J Vasc Surg 2011;54: Raffetto JD, Qiao X, Beauregard KG, Khalil RA. Estrogen receptormediated enhancement of venous relaxation in female rat: implications in sex-related differences in varicose veins. J Vasc Surg 2010;51: Raffetto JD, Qiao X, Beauregard KG, Tanbe AF, Kumar A, Mam V, et al. Functional adaptation of venous smooth muscle response to vasoconstriction in proximal, distal, and varix segments of varicose veins. J Vasc Surg 2010;51: Conte MS. Technical factors in lower-extremity vein bypass surgery: how can we improve outcomes? Semin Vasc Surg 2009;22: Raffetto JD, Khalil RA. Mechanisms of varicose vein formation: valve dysfunction and wall dilation. Phlebology 2008;23: Labropoulos N, Giannoukas AD, Delis K, Mansour MA, Kang SS, Nicolaides AN, et al. Where does venous reflux start? J Vasc Surg 1997;26: Bergan JJ, Schmid-Schönbein GW, Smith PD, Nicolaides AN, Boisseau MR, Eklof B. Chronic venous disease. N Engl J Med 2006;355: Kirsch D, Dienes HP, Küchle R, Duschner H, Wahl W, Böttger T, et al. Changes in the extracellular matrix of the vein wall the cause of primary varicosis? Vasa 2000;29: Kirsch D, Wahl W, Böttger T, Junginger T. [Primary varicose veins changes in the venous wall and elastic behavior]. [Article in German] Chirurg 2000;71:300-5; discussion Submitted Aug 8, 2011; accepted Oct 27, 2011.