NITROXYL ANION IN V ASORELAXATION: A POSSIBLE ENDOTHELIUM­

DERIVED HYPERPOLARIZING FACTOR AND V ASOPROTECTIVE

MOLECULE.

By

Brandi Michele Wynne

Submitted to the Faculty of the School of Graduate Studies of the Georgia Health

Sciences University in partial fulfillment of the Requirements of the Degree of Doctor of

Philosophy (Physiology)

2011 Nitroxyl anion in vasorelaxation: A possible endothelium-derived hyperpolarizing

factor and vasoprotective molecule.

This dissertation is submitted by Brandi Michele Wynne and has been examined and approved by an appointed committee of the faculty of the School of Graduate Studies of the Georgia Health Sciences University.

The signatures which appear below verify the fact that all required changes have been incorporated and that the dissertation has received final approval with reference to content, form and accuracy of presentation.

This dissertation is therefore in partial fulfillment of the requirements for the degree of Doctor of Philosophy.

Af>A \ 29 \2cJ 1\ Date Major Advisor It was observed that aorta from Angll hypertensive mice exhibit a preserved relaxation response to the HNO donor, AS, while mesenteric arteries exhibit a decreased AS­ mediated relaxation response. The role of voltage-gated potassium (Kv +) channels was explored as a possible mechanism for the decreased HNO vasorelaxation in the mesenteric arteries. And finally, using a model of ET-1 induced vascular dysfunction, a vasoprotective action of HNO was revealed.

These data demonstrate that NO- is a mediator of vasodilation, which is lost in the mesenteric arteries during Angll hypertension possibly through Kv + channel dysfunction.

This process may contribute to systemic hypertension. These data also reveal a therapeutic role for NO- donors, such as AS, which may lead to alternative treatment options during conditions where the use of organic nitrates are not effective. ACKNOWLEDGEMENTS

I would like to thank my dad, G. Rush Wynne, for dealing with all the questions from 'the others' relating to when I will finish school and for making me memorize the alphabet. It has come in handy over the years.

I am truly appreciative of all the help and support of my committee members;

Drs. Adviye Ergul, Jessica Filosa, David Fulton, Mong Wang and my three readers, Drs.

Alexis Stranahan, Nancy Kanagy and Richard White.

Huge amounts of gratitude are owed Dr. Rita Tostes, a previous committee member who morphed into a 'big science sister' combined with 'personal cheerleader'. I think that meeting you during my graduate career was divine providence.

. I could add an entire laundry list of friends whose help I have received ( and involuntarily ~nlisted) over the years; however, thesis paper is rather expensive and I would need many days and many pages to describe in detail. You know who you are, thank you.

And finally, copious quantities of gratitude are due to my mentor, Dr. R. Clinton

Webb. Although I am positive that he has read these exact same statements in the acknowledgements sections of his many students, I must say it again. Thank you for being an inspiration to all of us. Thank you for being so generous. Thank you for allowing me to pursue and develop my own interests. And thank you, ~or being such a wonderful mentor.

V TABLE OF CONTENTS

ACKNOWLEDGEMENTS ...... V

LIST OF FIGURES ...... ~ ...... viii

LIST OF TABLES ...... X

I. INTRODUCTION ...... 11

A. Statement of the problem and specific aims of the overall project...... 11

B. Brief literature review and discussion of the rationale of the project...... 3

Hypertension and Vascular Disease ...... 3

Vascular Smooth Muscle and the Endothelium ...... 4

Molecular mechanisms ofsmooth muscle relaxation and contraction ...... 4

Regulation ofrelaxation responses in the vasculature: role ofthe endothelium ...... 6

Nitroxyl Anion ...... 9

Production ...... 9

Physiology and pharmacology ofHNO ...... l 0

HNO as a vasoprotective molecule ...... 12

II. MANUSCRIPTS ...... 14

A. Nitroxyl anion mediates relaxation in mesenteric arteries from angiotensin II

hypertensive mice ...... 14

B. Aorta from angiotensin II hypertensive mice exhibit preserved nitroxyl anion

mediated relaxation responses ...... 3 8

VI C: Angeli's Salt, a nitroxyl anion donor, reverses endothelin-1 mediated vascular

dysfunction in aorta...... 60

III. UNPUBLISHED DATA ...... -...... 82

IV. DISCUSSION ...... 89

V. SUMMARY ...... 97

VI. REFERENCES ...... 99

APPENDIX ...... ~ ...... 107

Vascular smooth muscle cell signaling mechanisms for contraction to angiotensin II

and endothelin-1. Wynne, B. M., Chiao, Chin-Wei, Webb, R. C. Journal ofAmerican

Society ofHypertension. 2008 ...... 107

Mesenteric arteries from spontaneously hypertensive stroke prone rats exhibit an

increase in NO-dependent vasorelaxation via iNOS. Wynne, B. M., Giachini, F. R.,

Labazi, H., Lima, V. V., Carneiro, F., Dorrance, A. M., Webb, R. 'C., Tastes, R. C .. 107 LIST OF FIGURES

Figure 1. Chronic collection of MAP in conscience mice; sham and AngII-treated mice .

...... 31

Figure 2. Mesenteric arteries from AngII-hypertensive mice exhibit decreased ACh- and

AS-mediated relaxation responses ...... 32

Figure 3. Nitroxyl anion is a mediator in endothelium-dependent relaxation in mesenteric arteries from both sham and AngII-treated mice ...... 33

Figure 4. An inhibitor of voltage-gated potassium channels decreases relaxation in both sham and AngII-treated mice ...... 35

Figure 5. Relaxation responses to the nitroxyl anion donor, Angeli's Salt, are mediated through soluble guanylate cyclase ...... 36

Figure 6. Protein expression of calcitonin-gene related peptide is decreased and soluble guanylate cyclase ~1 increased in mesenteric arteries from AngII-treated mice ...... 37

Figure 7. Aorta from AngII-hypertensive mice exhibit endothelial dysfunction ...... 53

Figure 8. Angeli's Salt-mediated relaxation responses were preserved in intact and denuded aorta from AngII hypertensive mice ...... 54

Figure 9. Scavenging nitric oxide decreases ACh-mediated relaxation responses in aorta from AngII-treated and sham mice ...... 55

Figure 10. Nitroxyl anion does not primarily mediate endothelium dependent vasorelaxation in aorta from both sham and AngII-treated mice ...... 56

Figure 11. Voltage-gated potassium channel blockade decreases relaxation in aorta ...... 57

viii Figure 12. Relaxation responses to the nitroxyl anion donor, Angeli's Salt, are mediated through soluble guanylate cyclase ...... 58

Figure 13. Relaxation responses to the nitroxyl anion donor, Angeli's Salt, are reduced with nitric oxide and nitroxyl anion scavenging ...... 59

Figure 14·. Aorta incubated with endothelin-1 exhibit vascular dysfunction ...... 76

Figure 15. Incubation with AS reverses ET-1 induced vascular dysfunction in mouse aorta ...... 77

Figure 16. Incubation with the HNO donor, Angeli's Salt, reverses ET-1 mediated endothelial dysfunction ...... 78

Figure 17. Endothelin 1 treatment increases superoxide generation, which is decreased when co-incubated with Angeli's Salt...... 79

Figure 18. Incubation with the ONOO- degradation catalyst, FeTPPS, reduces ACh- mediated relaxation responses ...... _...... 80

Figure 19. Incubation with tempo I does not improve A Ch-mediated relaxation responses in Gtn and ET-1 treated aorta ...... 81

Figure 20. Angeli's Salt, the HNO donor, induces outward potassium current in aortic smooth muscle cells ...... 86

Figure 21. Angeli's Salt induces changes in current kinetics ...... 87

Figure 22. Concentration response curves to Angeli's Salt in mesenteric arteries induces relaxation through voltage-gated potassium channel activation ...... 88

Figure 24. Schematic of nitroxyl anion mediated relaxation via Angeli's Salt (a) and endothelium-mediated relaxation via acetylcholine ...... 95

Figure 25. Schematic for increased vascular resistance and hypertension ...... 96

lX LIST OF TABLES

Table 1 Pharmacological and physiological differences between NO' and NO-/HNO .... 13

X I. INTRODUCTION

A. Statement of the problem and specific aims of the overall project.

Hypertension is a leading cause of death and disability in the United States, and worldwide. In addition, hypertension is a risk factor for other cardiovascular diseases

(CVD) and stroke. The etiology of hypertension is varied and requires input from an array of factors; genetic, environmental and lifestyle. However, it is widely accepted that a change in vascular tone contributes to the pathogenesis of the disease.

Hypertension is characterized by endothelial dysfunction (ED), which is typified by a decrease in the relaxation response to vasodilatory agents, and an increase in the contractile response. A decrease in nitric oxide (NO) bioavailability and/or metabolism is a purported mechanism in this process. NO is a well-established endothelium-derived relaxing factor (EDRF) and aids in the maintenance of vascular tone, decreases platelet aggregation and can induce angiogenesis. Since its discovery, NO in the free radical form, has been considered to be the predominate mediator of vascular dilation in vivo.

However, the redox congener of NO, nitroxyl anion (NO), has emerged as an EDRF and a possible endothelium-derived hyperpolarizing factor (EDHF). The role of NO- in the vasculature is poorly understood and there are no studies investigating the role of this redox species in vessels, using a hypertensive model. Thus, three aims were established.

xi 2

Specific Aim 1: To test the hypothesis that during angiotensin II (AngII)-induced hypertension, vasodilation has an increased dependence upon HNO, and that this process may be mediated through activation of K+ v channels. To date, only basic studies regarding NO- in mesenteric arteries have been performed, with no studies determining whether these vessels demonstrate impairment in NO--mediated relaxation. In addition,

NO- has been suggested to be a new EDHF, and in this specific aim the role of potassium channels in NO--mediated relaxation was determined.

Specific Aim 2: To test the hypothesis that IINO-mediated relaxation will be preserved in aorta from Angll hypertensive mice. In the aorta, NO is considered to play a major role in vasodilation, and in vessels from AngII-hypertensive animals, decreased NO bioavailability is thought to be a primary cause for decreases in endothelium-mediated relaxation. Given that NO- is considered to be more stable than the free radical form,

NO', these studies were performed to determine if the aorta exhibits a similar decrease in

NO--mediated relaxation during Angll hypertension.

Specific Aim 3: To test the hypothesis that treatment with the NO donor, nitroglycerin

(Gtn), will exacerbate endothelin-1 induced vascular dysfunction via an increase in ROS, while treatment with the NO- donor, Angeli's Salt (AS) will not. Although Gtn is a widely used treatment for cardiovascular conditions, chronic administration of Gtn leads to tolerance and can induce vascular wall damage via reactivity with reactive oxygen species (ROS). NO- has been proposed to be more stable than NO', and administration of

NO-, using the donor AS, does not induce tolerance. 3

B. Brief literature review and discussion of the rationale of the project.

Hypertension and Vascular Disease

Hypertension is a leading cause of death and disability in the United States and globally. Although a variety of factors including, diabetes, smoking and other life style choices can be correlated with hypertension, there is still a subset of the population which becomes hypertensive due to unknown causes. These patients develop what is known as essential hypertension and frequently do not consistently respond to commonly used anti­ hypertensive medications.

The role of the kidneys and the nervous system in control of systemic blood pressure is important; however, vascular tone contributes to increased peripheral resistance and is known to be increased during chronic hypertension. Recent studies have

1 2 documented a profound role for vascuiar proteins in the maintenance of blood pressure. -

Michael and colleagues demonstrated that mice lacking the cyclic guanosine monophosphate ( cGMP)-dependent protein kinase PKGia exhibit significantly increased blood pressures. This is in the presence of completely normal renal hemodynamics and

1 2 salt handling. - In addition, investigators have demonstrated that increased blood pressure alters vascular contractility and relaxation through a variety of processes.

Altered production of contractile and relaxant factors by the endothelium is frequently referred to as endothelial dysfunction (ED). When there is a decrease in hyperpolarizing factors, a decrease in NO production and/or bioavailability and a propensity towards the production of endothelial-derived contracting factors (EDCF), then ED has occurred. This event has been determined to be a primary step in the 4 development of atherosclerosis and coronary artery disease, and is even a predictor of

3 8 cardiovascular disease (CVD). -

Vascular Smooth Muscle and the Endothelium

Arteries are composed of three layers; the tunica adventitia, the tunica media and the tunica intima. Although the outer and inner layers are composed mainly of connective tissue and endothelial cells, respectively, the tunia media is composed of primarily vascular smooth muscle.

The innermost layer, the endothelium, is an important modulator of vascular tone through the release of various vasoactive agents including nitric oxide (NO) and prostaglandins. Moreover, these agents are capable of activating different families of potassium (K+) channels, leading to vascular smooth muscle cell (VSMC) hyperpolarization and relaxation. Besides promoting VSMC relaxation, during conditions where the balance of relaxing and contracting factors becomes dysregulated, the endothelium becomes dysfunctional promoting a cycle of increased contraction and decreased relaxation.

VSMCs produce force through a process of cross-bridge cycling between actin and myosin filaments. In addition, VSMCs are capable of dynamic changes in gene expression resulting in differing phenotypes; from contractile to proliferative and secretory .1

Molecular mechanisms ofsmooth muscle relaxation and contraction.

Arterial lumen diameter is determined by the contractile state of VSMC, which in. tum respond to a variety of systemic and local physiological influences including sympathetic and neurohumoral, autocrine and paracrine factors (i.e. EDHF), local 5 chemical signals (i.e. pH) and other vasoactive hormones. 9 The contractile state of

VSMCs is controlled in part by intracellular calcium (Ca2+) concentrations and the level of myosin light chain (MLC) phosphorylation (pMLC). This phosphorylation event is necessary for actin and myosin interaction.

The myosin filament is composed of thousands of myosin molecues, which are interspersed between the actin filaments. Actin filaments (F-actin) are large, polymerized protein molecules which are arranged within a smooth muscle cell (SMC) from a central point, referred to as a dense body. This arrangement of dense bodies within the SMC allows for the range of contractile force seen in SMC. MLC is a 20 kilodalton (kDa) chain of the myosin (heavy) chain and initiator of contraction.

Two enzymes regulate the phosphorylation status of MLC at serine 19 (serl 9), myosin light chain kinase (MLCK) and myosin light chain phosphatase (MLCP). MLCK becomes activated through an increase in intracellular ca2+, which associates with

2 calmodulin forming the Ca +-calmodulin complex, which subsequently interacts with

MLCK and leads to activation. MLCK phosphorylates MLC, allowing for association with actin and cross-bridge cycling via the energy released from adenosine triphosphate

(ATP) hydrolysis. This process occurs because of the intrinsic adenosine- 5' - triphosphate (ATP)-ase activity of the myosin head. Relaxation occurs when MLC is de­ phosphorylated, an event which occurs through the action of MLCP.

2 VSMC also produce and maintain force generation through Ca +-independent mechanisms. The Rho-kinase (ROK, p160ROCK) pathway is a major effector of MLCP activity. ROK mediates contraction by means of altering MLCP phosphorylation.

MLCP is regulated through the catalytic 3 8-kDa head of the type 1 protein phosphatase 6

(PP 1c) subunit. Rho A is a small, monomeric G protein which is regulated (and activated) through the binding of guanosine triphosphate (GTP), and is facilitated through the Rho-guanine exchange factors (RhoGEFs). With the activation of RhoA, this RhoA­

GTP complex then activates ROK. This serine/threonine kinase phosphorylates the myosin-binding subunit of MLCP, thus inactivating it and allowing for a sustained contraction.

Protein kinase C (PKC) is also an important mediator of VSMC contraction, and can be Ca2+-dependent or -independent. Diacylglycerol (DAG) is a second messenger which is activated through G-protein coupled receptor (GPCR) ligand binding and activation of phospholipase C beta (PLCp). PLCP is responsible for the cleavage of phosphoinositide 4, 5-bisphosphate, which results in inositol trisphosphate (IP3) and

DAG. IP3 and DAG regulate two distinct, but parallel pathways culminating in the phosphorylation of MLC. DAG activates the serine/threonine kinase PKC, which can alter CPI-17 and ROK activity; both lead to MLCP inactivation. In addition, PKC can directly activate MLCK. IP3 regulates Ca2+ flux channels in the sarcoplasmic reticulum

1 10 11 (SR), which also leads to increased contraction. ' -

Regulation ofrelaxation responses in the vasculature: role ofthe endothelium.

In 1980, Furchgott and colleagues revealed that removal of the endothelium left the arteries resistant to ac~tylcholine (ACh)-mediated vasodilation. 12 In this seminal work, it was determined that the endothelium releases factors which promote smooth muscle relaxation and the term endothelium derived relaxation factor (EDRF) was

6 12 13 coined. ' • In the same decade investigators found that a simple gas, NO, was the 7

14 18 purported EDRF. - Since this discovery, researchers have uncovered many other

EDRFs and EDHFs which also regulate vascular tone. Many of these factors act via gap junctions and in an NO-independent mechanism, lead to VSM hyperpolarization.

Although much of the endothelium-dependent relaxation in response to neurohumoral mediators and physical forces can be attributed to either the release of NO and/or prostacyclin (PGh), there are still other factors which induce VSM relaxation.

NO, the free radical gas is a potent protector of the vasculature but can have deleterious effects if produced in copious quantities or when deficient. There are three nitric oxide synthase isoforms, from three genes each of which are located on different chromosomes; neuronal NOS (nNOS, also termed as NOS 1), inducible NOS (iNOS, also termed NOS2), and endothelial NOS (eNOS, also termed NOS3). Under physiological conditions, NO is produced through the 5-electron oxidation of L-arginine to L-citrulline;

19 requiring tetrahydrobiopterin (BH4) and molecular oxygen. In the absence of cofactors,

NOS uncoupling occurs leading to decreased NO production and increased reactive oxygen species (ROS) generation.20

Each of these isoforms performs a specific function. Endothelial NOS is a constitutive, cytosolic enzyme, only producing small amounts of NO in response to receptor or physical stimulation. On the other hand iNOS is activated in macrophages, endothelial or other cell types following feed-forward cytokine stimulation, such as interleukin-I~, interferon-y and tumor necrosis factors. 21 nNOS is an important modulator of non-adrenergic, non-cholinergic neurotransmission in tissues such as the corpus cavemosum and airway smooth muscle. 8

In addition to its important role as a second messenger, NO inhibits platelet aggregation and reduces the expression of cell surface adhesion molecules on the

19 22 endothelium. • With the lack of a healthy, intact endothelium, any damage to the vessels results in a feed forward system of coagulation and platelet activation resulting in the release of constricting agents further promoting hemostasis. 6

The family of cycloxygenase (COX)-derived prostaglandins serves a two-fold function in the vasculature depending upon the functionality of the endothelium. With 2 agonist stimulation and an increase in intracellular Ca + and phospholipase A2 cleaves arachadonic acid (AA). This process leads to prostaglandin synthase (PGIS) synthesizing a major EDRF, prostacyclin (PGh). PGh induces VSM relaxation through activation of the PGh receptor, IP and downstream cyclic adenosine monophosphate (cAMP) accumulation. Interestingly, this major vasorelaxant is also a EDCF during conditions where the TP receptor ·predominates, such as hypertension and other cardiovascular 5 conditions. In addition, COX over-activation can lead to endoperoxidase (PGH2) production and further activation of the TP rec·eptor which directly induces increases in

2 3 5 23 VSM intracellular Ca + and contraction. · •

However, NO and PGh are only two important vasodilatory agents released from the endothelium. Many such relaxation responses occur in the presence of NOS and

COX inhibitors, such as L-NAME and indomethacin, resp., further confirming the

6 24 25 presence of a growing number of EDRF and EDHF-like agents. • - These include the epoxyeicosatrienoic acids (EETs), hydrogen peroxide (H2O2), hydrogen sulfide (H2S) and the redox variant of NO, nitroxyl anion (NO} 9

Nitroxyl Anion

Since the discovery of NO as the putative EDRF, the vast majority of research has focused on the free radical congener of this gas. However, nitrogen monoxide also exists

26 28 as the nitrosonium cation (NO+) and as a one electron reduced species, N0-. - Dr.

Angelo Angeli, an Italian chemist, was a pioneer in his field and his work on pyrrole and indolic compounds was considered a breakthrough at the time. Over 100 years ago, he

also discovered that "salts of nitroxylaminic acid are readily resolved into the

1 29 corresponding nitrites and the unsaturated nitroxyl"( ). For many years Angeli's Salt

(AS), was used bench-side as a nitroxyl anion donor, but the entity of NO- was relegated

'un-physiological' due to the incorrect assumption that the pKa was 4.7, leaving it an

anion in vivo. 30 In 2002 this error was corrected. Investigators found that the pKa was

greater than 11, suggesting that in vivo, NO- would not only exist protonated' (HNO), but

31 33 more stable at a physiological pH. - Since this discovery, the biochemistry, pharmacology and physiology of HNO has emerged revealing HNO a distinct entity from that of NO.

Production

HNO and NO can both be produced through the conversion of L-arginine to NO

by nitric oxide synthase; however, HNO is particularly produced when NOS is uncoupled

or in the absence of the NOS cofactor, tetrahydrobipterin (BH4) through direct oxidation

of NOS intermediates, .N°-hydroxy-D-arginine and hydroxylamine. Interestingly, hydroxylamine has been detected in plasma and some cells, which suggests that this may

34 35 37 be a possible endogenous pathway for HNO production. , - While NO can be stored

1 Palocci, N. et al. AJP: Heart Cir Physiol. 2009 pg: 10 as nitrites and nitrates and thus released from tissues, HNO rapidly dimerizes and is 38 39 converted to nitrous oxide (N2O). - In contrast, HNO has been shown and hypothesized to be produced from a variety of other sources. Although unlikely, HNO can be directly generated from NO by mitochondrial cytochrome c (mCyC), xanthine oxidase (XO), ubiquinol, haemoglobin and manganese superoxide dismutase

(MnSOD).28-29' 394° Furthermore, the highly thiophillic nature of HNO would suggest a direct interaction with S-nitrosothiols. Indeed, Cheong and colleagues found that HNO directly stimulates the ryanodine receptor (RyR), possibly through S-nitrosylation

(SN0).41-42 43

Physiology and pharmacology ofHNO

Although the definitive endogenous source for HNO remains yet determined, the chemistry of HNO would allow it to diffuse directly across membranes, with no need for channel or po.re. High levels of glutathione (GSH) exits in the intracellular space of the cell; thus, given the distinct thiophillic chemistry of HNO as compared to NO, this would suggest that HNO may be directed towards membrane-bound targets, such as enzymes, channels and receptors for possible downstream _targets. 32, 44-45 ·

The mechanism by which HNO elicits relaxation varies from NO. Although still controversial, there is evidence· for both NO and HNO to activate soluble guanylate cyclase (sGC), and using in vitro techniques, cyclic guanosine monophosphate (cGMP) seems to accumulate in large conduit vessels only.29' 39, 46 Seminal work published by

Fukuto, et al. in 1992 demonstrated that HNO could induce sGC-dependent vasorelaxation in rabbit aorta and bovine intrapulmonary artery.29' 47 Given that much of the relaxation response of AS, the HNO donor, can be inhibited with the use of ODQ (a 11 sGC inhibitor) it has been the subject of more than one conflicting paper. Biochemical studies suggest that reaction of a ferric heme with HNO would actually lead to an equivalent species of the reaction of ferrous heme with NO. This suggests that HNO would possibly have similar activity with the ferric heme of sGC that NO has with ferrous heme under normal conditions. However, the ferric form of sGC has been shown to be HNO-resistant by some investigators, but preferentially activated by HNO by others

48 51 in the field. - This is complicated by a recent article demonstrating that HNO can only activate ferrous-coupled sGC. 52 It is interesting to note that the activation of HNO, in vivo, by directly giving AS does not lead to the accumulation of plasma cGMP, in contrast to NO donors, such as sodium nitroprusside (SNP). If HNO does preferentially bind ferric (Fe3+) haem, then this would truly be interesting as ferric haem is suggested to

29 39 46 be the NO-insensitive sGC isoform and may predominate in disease. ' , HNO, while activating sGC, also leads to the production of calcitonin gene-related peptide (CGRP), which is a potent vasodilator released from neurons and has been determined as a

39 53 biomarker of HNO activity. ' CGRP activation by HNO has even been proposed as a mechanism to account for the variation in responses seen in the capacitance vessels.

As far as relaxation responses in small resistance arteries, HNO elicits vasodilation differently than that of NO. Furthermore, investigators in the field have also established that HNO activates K+ channels in mesenteric and coronary vasculature; voltage-gated K+ (K+ v) in rat mesenteric and ATP-sensitive K+ channels (KATP) in rat

54 56 coronary arteries. - This is in direct contradiction with NO, which has been shown to activate only Ca2+-activated potassium (Kea) channels, such as BK channels. Recently, whole cell patch clamp recordings demonstrated a definitive mechanism for HNO- 12 induced hyperpolarization of VSMCs as well as inducing spreading hyperpolarization and dilations using an in vitro preparation. 55· 57

HNO as a vasoprotective molecule

The use of HNO donors clinically is not without precedent. The HNO donor, cyanamide, has been used classically to treat conditions such as chronic alcoholism via

SNO of N-methyl-D-aspartate (NMDA) receptors.40 However, recent studies suggest that

HNO donors confer protection during ischemia-reperfusion injury, similar to that of

NO. 58-59 In addition, HNO donors have been shown to induce increased lusitropy and inotropy of cardiac tissue, thus decreasing preload and increasing ejection fraction. This may prove HNO donors to be an important therapeutic agent for the treatment of heart failure.4o, 60

Moreover, the use of HNO donors in vivo and in vitro, does not develop tolerance or cross-tolerance when given acute or chronically. This is in stark contrast to other organic nitrates.61 -63 The development of tolerance is a well established difficulty with clinical use of organic nitrates, such as nitroglycerin (Gtn). Two main mechanisms of nitrate tolerance are under investigation, diminished bioconversion of Gtn and increased

ROS production. HNO confers greater stability and less reactivity with reactive oxygen 64 39 65 species as compared to NO; especially with regards to superoxide (02-). · This would suggest that HNO donors would be less reactive with already present vascular

ROS as compared to NO donors. Additionally, HNO has been demonstrated to directly inhibit aldehyde dehydrogenase (ALDH2), a regulatory enzyme in the bioconversion of

Gtn to NO_ 39-4o, 66 This may establish HNO as a preferred vasodilator over other NO donors. 13

Table 1 Pharmacological and physiological differences between NO· and NO-/HNO.

Properties NO- Nff/HNO

Possible endogenous sources NOS, coupled NOS, coupled NOS, uncoupled NOS intermediate N11-hydroxy-I arginine S-nitrosothiols S-nitrosothiols Intracellular reduction of NO·*

Thiol reactivity Does not directly react Highly thiophillic

Vascular smooth muscle signaling sGC/cGMP mediated sGC/cGMP* (Fe2+-heme) ( controversial) 1~ii_tf::,e~s_:_••··-- ...... _....._"""---...... _,_,• Resistance vessels ( mesenteric)

Cardiac inotropic effects Negligible or negative Positive; induces lusitropy and inotropy

Responses scavenged by Carboxy-PTIO L-cysteine Hydroxocobalamin N-acety 1-L-cysteine Diothiothreitol

References are included in text. *indicates highly controversial area; data inconclusive and/or contradictory 14

II. MANUSCRIPTS

A. Nitroxyl anion mediates relaxation in mesenteric arteries from angiotensin II

hypertensive mice.

ABSTRACT:

Aims: Nitroxyl anion (NO) has recently become an emerging candidate in vascular

regulation. NO- is a potent vasodilator of both conduit and small resistance vessels and

mediates relaxation in a soluble guanylate cyclase-dependent manner. Interestingly, NO­ activates voltage-dependent K+ (K+v) channels, whereas nitric oxide (NO) activates

calcium-activated K+ channels. To date, there are no studies investigating the role of NO­

in hypertension, and the possible mechanisms which may be altered during this condition.

We hypothesized that mesenteric arteries from angiotensin II-induced (AngII)

hypertensive mice would exhibit an increased dependence upon NO- for relaxation,

which may be mediated through K+ v channels.

Methods and Key Results: C57/Bl6 mice, aged 12-14 weeks were implanted with mini­

pumps containing angiotensin II (AngII, 90ng/min) for 14 days. Mice were given high

salt (4%, HS) chow. First order mesenteric arteries were isolated and used in functional

studies or frozen for Western blot analysis. We observed that mesenteric arteries from

AngII mice (AngII) exhibited a decrease in HNO-mediated relaxation, which was

endothelium-independent. With HNO scavenging by L-cysteine [3mM], the maximal 15 acetylcholine (ACh)-mediated relaxation response was decreased in sham, whereas mesenteric arteries from Angil exhibited a decrease in sensitivity. Incubation with the

K+ v channel inhibitor, 4-aminopyridine [lmM], decreased ACh-mediated relaxation responses in sham, but almost completely abolished relaxation in Angil.

Conclusions: We reveal that HNO-mediated relaxation is impaired in mesenteric arteries from Angil treated mice, and that K+ v channel activation may be a cause for this impairment.

Introduction

The small arteries or arterioles, known as the resistance vessels, contribute to systemic pressure through increased contraction and/or relaxation of the vascular smooth muscle (VSM) within the tunica media. These changes in vessel lumen diameter are mediated through a variety of circulating and locally released agents and neurogenic stimuli.

The endothelium, which lines the lumen of all vessels, performs various functions including influencing vascular tone through a tightly balanced system of dilators and constrictors. Vasoconstrictors, such as thromboxane and endothelium-derived contracting factor (EDCF) and physical forces, elicit contraction while vasodilators, such as nitric oxide (NO), prostacyclins and EDHF produce relaxation. 23 The balance between vasodilators and vasoconstrictors is crucial for maintenance blood flow and pressure ( for

5 a thorough review of EDCF, see Vanhoutte et al. ). 16

Although there are a variety of factors which cause relaxation in resistance arteries, EDHFs contribute significantly to this process. EDHF was conventionally thought to be one factor but can be considered a group of factors which elicit relaxation by hyperpolarization of the VSM. 6' 24' 67-68 Recent evidence suggests that a redox variant of NO, the nitroxyl anion (NO), may mediate relaxation through potassium (K+) channel activation, leading investigators to suggest that NO- is an EDHF. 25

NO- is the one-electron reduction product of NO, and until recently was considered to be non-physiological, an entity only present in the laboratory due to the incorrect assumption that NO- exists as an anion (pKa 4.7). 30 However, this assumption was corrected in 2002, when investigators determined the actual pKa to be greater than

11, suggesting that at a physiological pH, NO- exists as HNO. 31 -32 Since this discovery, the role of HNO has been intensively studied. The physiology, pharmacology and biochemistry of HNO is vastly different than that of NO.

HNO and NO can both be produced through the conversion of L-arginine to NO by nitric oxide synthase (NOS); however, HNO is mainly produced when NOS is 35 37 uncoupled or in the absence of the NOS cofactor, tetrahydrobipterin (BH4). - This is of particular interest when one considers that during pathophysiological conditions, such as hypertension and, increased inflammation, BH4 can become oxidized and NOS uncoupling occurs with the increase in reactive oxygen species (ROS). 69 During these same conditions, the product of uncoupled NOS, N-hydroxyl-L-arginine, can lead to

HNO production. 7° Furthermore, HNO can be produced through the reduction of NO by mitochondrial cytochrome c (mCyC) and non-enzymatically from the reaction of S­ nitrosothiols with thiols. 41 ' 56' 71 17

The mechanism by which HNO elicits relaxation varies from NO. Both activate soluble guanylate cyclase (sGC); however, while NO binds to the ferrous (Fe2+)-heme moieties of sGC, HNO preferentially binds ferric (Fe3+), which is the NO-insensitive sGC

29 39 46 isoform which predominates in disease. • • HNO, while activating sGC, also leads to the production of calcitonin gene-related peptide, which is a potent vasodilator released from neurons and can be used as a biomarker of HNO activity. 39, 53

Interestingly, investigators in the field have also established that HNO activates K+ channels in mesenteric and coronary vasculature; voltage-gated K+ (K+ v) in rat

54 56 mesenteric and ATP-sensitive K+ channels (KATP) in rat coronary arteries. -

Although there is an increasing body of knowledge regarding HNO-mediated relaxation, no current studies have investigated the relative role that HNO plays during hypertension. Given the different biochemical properties of HNO as compared to NO,

HNO may mediate EDHF-like effects previously attributed to NO. We hypothesize that during angiotensin II (Angll)-induced hypertension, vasodilation has an increased dependence upon HNO, and that this process may be mediated through activation of K+ v channels.

MATERIALS AND METHODS

Male C57bl/6 mice, weighing between 25-30 grams were obtained from Jackson

Laborotories (Bar Harbor, ME). Mice were maintained on a 12-hour light dark cycle, housed five per cage and allowed access to chow and water ad libitum. All procedures were performed in accordance with the Guiding Principles in the Care and Use of 18

Animals, approved by the Medical College of Georgia Committee on the Use of Animals

in Research and Education.

Blood Pressure Telemetry Studies

Mice were anesthetized using isofluorane and a DSI Data Transmitter (Data

Sciences International, St. Paul, MN) was implanted in the left carotid artery, routed and

secured sub-scapularly. Animals were allowed to recover; systolic/diastolic pressures, heart rate and activity were collected for 18 hours per day/night. Data were analyzed

using Power Lab (AD Instruments, Colorado Springs, CO). The blood pressure data

from these mice are shared between two publications.

Functional Studies

After euthanasia with CO2, the mesentery was rapidly excised and bathed in ice­

cold physiological salt solution (PSS) (NaCl 120 mM, KCl 4.7 mM, KH2PO4 1.18 mM,

NaHCO3 14.9 mM, dextrose 5.6 mM, CaChH2O, 0.06 mM EDTA). Increased

concentrations of EDTA were used in PSS buffer to aid in preventing the extracellular

conversion of HNO to NO. First order mesenteric vessels were carefully isolated and

mounted as ring preparations on two stainless steel wires (= 200 mm in length with

internal diameter = 100 to 150 µm) in a small-vessel myograph (Danish MyoTech,

Aarhus, Denmark). Vessels were maintained at 37° C and continuously aerated with 95%

02, 5% CO2 and allowed to stabilize for at least 45 mins, at an optimal passive force of

2.5 mN. After stabilization, tissues were contracted with KCl (120 mM) solution to

determine the reactivity of the vascular smooth muscle cells. To determine the viability

of the endothelium, contraction was stimulated via phenylephrine (Phe; 10 µM) followed 19 by relaxation induced by acetylcholine (ACh; 10 µM). Vessels were then washed, as described previously, before performing concentration response curves (CRC) and after each CRC. CRCs to ACh or Angeli's Salt (AS, Cayman Chemical, Ann Arbor, MI) in

Phe-contracted vessels was performed in the presence ·of vehicle or the following inhibitors: carboxy-PTIO (CPTIO, nitric oxide scavenger), L-cystiene (L-cys, nitroxyl anion scavenger), 4-aminopyridine (4-AP, K+v channel blocker). CPTIO was obtained from Cayman Chemical and 4-AP obtained from Tocris Bioscience, Ellisville, MO. All other chemicals and drugs were purchased from Sigma Aldrich, St. Louis, MO. Force measurements were collected using Chart™ Software (ADI Instruments, Colorado

Springs, CO) for PowerLab data acquisition systems (ADI Instruments).

Tissue homogenization

Following dissection and fat removal, mesenteric arteries were quick frozen with liquid nitrogen, pulverized in a liquid nitrogen-cooled mortar and pestle and solubilized in a 255 mM sucrose/I 0mM Tris buffer (pH 7.4) with protease inhibitors (0.5 mM

PMSF, 2 mM EGTA, 10 µg/ul aprotinin and 10 µg/ml leupeptin) and tyrosine phosphatase (1 mM sodium orthovanadate) inhibitors. Homogenates were centrifuged

(3000g for 1 hour, 4°C) and supernatant total protein (Bio-Rad~ Hercules, CA) measured.

Western Blotting

Total supernatant protein ( 40 µg total, 4: 1 in denaturing sample buffer and boiled for 5 mins), was separated on a 10% SDS-polyacrylamide gel (SDS-PAGE) and transferred to Immobilon-P (Bio-Rad) nitrocellulose membrane. Membranes were blocked to rule out non-specific binding using 5% skim milk in Tris-buffered saline 20

solution with Tween (0.1%) for 1 hour at 24°C. Membranes were then probed overnight

( 4°C) with primary antibodies as follo)Vs: CORP (1: 100), sGCa1 (1: 1000). All antibodies

were purchased from Cell Signaling, Danvers, MA. Blots were washed three times with

TBS-T (30:5:5 mins) and once with TBS (5 mins). Depending upon the antibody, either

· anti-rabbit or anti-mouse horseradish peroxidase -linked secondary 1antibody (Amersham

Labs, Piscataway, NF) was added for one hour and incubated with the blots at 4°C. Blots

were washed again using the described regimen. Enhanced chemiluminescence (Super

Signals Ultra, Pierce, Rockford, IL) was used to visualize labeled bands. Densitometry

was evaluated using Un-Scan-It image densitizing program (Orem, UT).

Statistical Analysis

Agonist concentration-response curves were fitted using a nonlinear interactive

fitting program (GraphPad Prism, Graph Pad Software Inc., San Diego CA), and values

expressed as percent of maximal relaxation graphed against increasing molar

concentrations of agonist. Agonist potencies and maximum response are expressed as

negative logarithm of the molar concentration of agonist producing 50% of the maximum

response (EC50) and maximum effect elicited by the agonist (Rmax), respectively. Non­

linear regression analysis was used to determine EC5o values, where Rmax was normalized

to 100 percent for calculations. Data are expressed as mean±SEM (n), where n is the

number of experiments performed. Statistical analysis of the concentration-response

curves (CRS) was performed by using the F test for comparisons of best-fit data between

groups (EC50 and Rmax), For CRCs where a best-fit analysis could not be performed,

absolute Rmax values were averaged ,and Students' t-test performed for significance. 21

Western blot data were densitized using the average of protein expression in each gel normalized to ~-actin. Data from multiple gels are expressed as fold change

(arbitrary units, AU) of sham. Values were analyzed for statistical significance using

Students' t-test for comparison between all groups. Values of p<0.05 were considered a statistically significant difference.

Drugs

Carboxy-PTIO was suspended in DMSO and Angeli's Salt was prepared using a

0.0lM NaOH solution. All other stock solutions were prepared by using water. Stock solutions originally diluted in DMSO or ethanol were used with a final concentration of less than 0.003% v/v in the muscle bath; this concentration has been demonstrated to have no effect on vascular reactivity. Additionally, solutions containing vehicle levels of ethanol and DMSO were also used through the experimental protocol.

RESULTS

Angiotensin II treated mice exhibited an increase in mean arterial pressure, as shown by telemetry.

Before drug treatment, MAP, HR and activity data were obtained during a control period of no less than 4 days in all mice (Figure 1). Mice treated with Angll for two weeks exhibited a significant increase in MAP by day 5 ( 151 ± 5 mmHg) as compared to sham surgery mice (114 ± 0.7 mmHg, p<0.01).

Ang/I-hypertensive mice exhibit a decreased relaxation response to ACh and to the nitroxyl anion donor, Angeli's Salt. 22

Angil hypertensive mice exhibited a significantly decreased relaxation response to ACh when compared to sham vessels (EC50 -5.73±0.16 vs. -6.33±0.08, p<0.0001). No differences in maximal relaxation were observed (Figure 2a).

As shown in Figure 2b, mesenteric arteries from AngII hypertensive mice demonstrated a decreased sensitivity to AS (EC50 -5.34±0.13 vs. -6.28±0.18, p<0.001).

There were no differences observed when these vessels were denuded (EC5o -5.31±0.23 vs. -6.42±0.14, p<0.01), demonstrating that this phenomenon is endothelium-independent

(Figure 2).

Mesenteric arteries from Angil hypertensive mice are dependent upon nitroxyl anion for relaxation.

To further investigate the role of HNO in vasorelaxation, scavengers of NO and

HNO (CPTIO and L-cys, resp.) were used. When vessels from AngII-treated mice were incubated with CPTIO, a NO scavenger, no differences in sensitivity (EC50) or maximal relaxation (Emax) were observed (Figure 3a).

In Figure 3b, AngII hypertensive and sham mesenteric arteries were incubated with the HNO scavenger, L-cys. Previous investigators have demonstrated that L-cys is

54 72 capable of effectively scavenging HNO biochemically and in ex vivo preparations. , -

73 Sham vessels incubated with L-cys exhibited a decrease in the maximal ACh mediated relaxation response (Rmax 68.7% ± 5.0 vs. 94.4% ± 3.3, p<0.001). In contrast, mesenteric arteries from AngII hypertensive mice exhibited a significant reduction in ACh sensitivity (ECso -5.08 ± 0.23 vs. -5.73 ± 0.16, p<0.05). When vessels were incubated with scavengers of both NO and HNO (Figure 3c), no additive effect was observed. 23

Voltage gated potassium channels play a regulatory role in ACh-mediated vasorelation

in mesenteric arteries from Angll hypertensive mice.

K+ v channels are known to be differentially activated in mesenteric arteries via

74 76 HNO versus NO, which activates K+ Ca channels. - Given this, the role of K+ y channels in this model of hypertension was investigated. Mesenteric arteries were incubated with 4-AP, which has been previously demonstrated to be a specific K+ v

25 54 channel blocker. ' As shown in Figure 4, mesenteric arteries from sham and Angil­ treated mice incubated with 4-AP revealed a decrease in ACh-mediated vasorelaxation.

Sham vessels with K+ v channel blocker, as compared to vehicle, exhibited a decreased sensitivity (EC50 -5.68 ± 0.14 vs.-6.53 ± 0.08, p<0.0001), but there was no decrease in maximal relaxation. In hypertensive vessels, we observed an almost complete inhibition

of relaxation, with a maximal relaxation of 14.6% ± 4.8. These data-suggest a major role for K+ v channels in relaxation during Angil hypertension. Other published data also

suggest that these channels are not only important during HNO-mediated relaxation, but activated distinctly by HNO in resistance vessels, as compared to other K+ v channels. 54

AS-mediated relaxation is dependent upon both K/ channels and soluble guanylate cyclase.

To aid in determining the mechanism involved in HNO-mediated relaxation, CRC to AS were performed in mesenteric arteries from sham and Angil-treated mice. Vessels

were incubated with 4-AP, the Kv+ channel blocker, as shown in Figure 5a. Interestingly,

we observed no significant differences in sensitivity to AS in either sham or Angil­

treated vessels. Additionally, Kv + channel blockade did not reduce the AS-mediated 24 relaxation response. In vessels incubated with ODQ, the sGC inhibitor (Figure Sb), both sham (Rmax 27.0%) and Angil (Rmax 11.5%) vessels exhibited a decrease in the maximal relaxation response to AS.

CGRP protein expression is decreased in mesentery from Angll-treated as compared to sham.

The mechanisms of HNO mediated relaxation are still poorly understood. It is accepted that a portion of both NO, and HNO vasorelaxation is sGC-dependent.

However, although reports show that cGMP is increased during infusion of NO donors, controversy remains when HNO donors, such as Angeli's Salt, are given. It is known that HNO induces an increase in levels of the neurodmodulator CGRP and although no mechanism has been suggested, CGRP is currently accepted as a biomarker for HNO

53 56 58 production. , ' Western blot analysis was used to determine CGRP levels in sham,

Angil mesentery. We observed that CGRP levels were decreased by half in Angil mesentery (Figure 6a, p<0.05) as compared to measures in sham mice.

Protein expression ofsGC subunit p1 is increased in Angll mesentery.

Functional data from this study demonstrate that Angil hypertensive mesenteric arteries are dependent upon HNO for ACh-mediated relaxation. To investigate a mechanism for this response, protein expression of sGC (subunit ~1) was evaluated.

Protein levels of the ~1 subunit were found to be increased in mesenteric arteries from

Angil mice as compared to those from sham. (Figure 6b, p<0.05). These data suggest a compensatory mechanism by Angil hypertensive vessels to increase NO-mediated signaling in the vasculature. 25

DISCUSSION

A newly emerging modulator of vascular tone is the one-electron reduced product of the well known vasodilator NO, called HNO. Although previously it was considered to only be an artifact in the laboratory, with up-to-date testing proving that NO- exists as a protonated entitity, new studies regarding this molecule in the vasculature have emerged. Even so, others have argued that HNO is at best, an intermediate to NO production. In a paper published by Favalaro and colleagues in 2009, they measured reactivity in resistance arteries coupled with single cell recording of membrane potential

(mV). In those studies, they demonstrate that membrane hyper-polarization occurs concurrently with relaxation to AS in mesenteric vessels. 54 This study gives overwhelming support to the hypothesis that HNO mediates vasorelaxation in vivo, and through an EDHF-type mechanism of postassium channel activation and hyperpolarization. In the present study we corroborate these observations, and other previous findings demonstrating a role for HNO in endothelium-mediated vasorelaxation.

It is widely accepted that during hypertension vascular and endothelial dysfunction is present, due to changes in contraction and structural properties of the endothelium and VSMC. 77 Although changes in the endothelium lead to alterations in receptors and endothelium-derived contracting factors (EDCFs), endothelial dysfunction

5 24 can be characterized by a decrease in the ability of the vasculature to relax. • In this study, a model of Angll hypertension was used. Angll modulates vascular contraction through a variety of signaling pathways. Additionally, increased levels of Angll are known to contribute to increased vascular resistance, smooth muscle cell proliferation and increased production of ROS. It has been demonstrated that Angll induces increased 26

1 78 expression of NADPH oxidases, contributing to ROS production. ' Studies have also demonstrated a significant increase in circulating levels of pro-inflammatory cytokines during several models of salt-sensitive hypertension, including Angll-dependent hypertension. AngII can modulate pro-inflammatory gene transcription and activate a

79 81 variety of inflammatory cell signaling mechanisms within cells. -

In first order mesenteric arteries from Angll-treated mice, there was a significant decrease in ACh-mediated relaxation responses (Figure 2a). Additionally, we observed that there was an endothelium-independent decrease in the relaxation response to the

HNO donor, AS (Figure 2b). Although endothelium dysfunction was indeed present in conduit (unpublished observations) and mesenteric arteries from Angil-hypertensive mice, only the mesenteric arteries exhibited HNO-mediated dysfunction. As mentioned previously, MAP was measured in the Angil-treated mice and we observed a significant increase in MAP by day 3 (Figure 1).

Since both vasodilators, NO and HNO, activate sGC, we measured protein levels of sGC in the mesenteric arteries. In addition, our data suggests that the HNO donor, AS, mediates relaxation in a sGC-dependent manner. We found that the Angil-treated mice express significantly increased protein levels of sGC (Figure 6b). This may contribute to the differences in AS sensitivity, as our data show that AS is predominantly· sGC dependent, and most of the relaxation response to AS could be abolished with the use of the sGC inhibitor ODQ (Figure Sb). Literature suggests that sGC becomes NO-resistant during conditions such as hypertension, by being converted to the ferric form when there are 'oxidizing conditions' .49 The increase in sGC protein expression that we observed in 27 the Angll mesenteric arteries may be a compensatory mechanism to increase vasodilation. Seminal work published by Fukuto, et al. in 1992 demonstrated that HNO could induce sGC-dependent vasorelaxation in rabbit aorta and bovine intrapulmonary

9 47 artery.2 ' Given that much of the relaxation response of AS, the HNO donor, can be inhibited with the use of ODQ a sGC inhibitor, it has been the subject of more than one conflicting paper. Biochemical studies suggest that reaction of a ferric heme with HNO would actually lead to an equivalent species of the reaction of ferrous heme with NO.

This suggests that HNO would possibly have similar activity with the ferric heme of sGC that NO has with ferrous heme under normal conditions. However, the ferric form of sGC has been shown to be HNO-resistant by some investigators, but preferentially

48 51 activated by HNO by others in the field. - This is complicated by a recent article

52 demonstrating that HNO CcJ.n only activate ferrous-coupled sGC. It is interesting to note ·that the activation of HNO, in vivo, by directly giving AS does not lead to the accumulation of plasma cGMP, in contrast to NO donors, such as sodium nitroprusside

(SNP). Overall, this suggests the need for further investigations into the role of sGC activation by HNO, not only purely for physiological importance, but also for therapeutic options.

Similar studies were then performed using the endothelium-dependent vasodilator, ACh to compare to CRCs performed to AS. Using scavengers for both NO

(CPTIO) and HNO (L-cys) we were able to differentiate vasorelaxation of these two molecules. Using the NO scavenger, we found that reducing NO bioavailability led to a small decrease in the maximal relaxation in mesenteric arteries from sham and a negligible decrease in vessels from Angll mice. Given the highly thiophillic nature of 28

HNO, it has been used as a scavenger by many investigators and numerous papers have

_been published demonstrating the ability of L-cysteine (L-cys) to differentiate between

No and HNO b10c. h em1ca. II y. 54-55 ' 72-73 ' 82-83 In our vascuI ar react1v1ty· · stud' 1es, we 1oun.(:'. d that L-cys decreased the maximal ACh-mediated relaxation response in sham mice and that L-cys incubation in vessels from Angil-treated mice also exhibited a significant rightward shift in the CRC (Figure 3b). These data suggest a role for HNO in ACh­ mediated relaxation; and given that this is an endothelium-dependent process, supports other studies demonstrating that HN O is an in vivo mediator of vasorelaxation.

As mentioned previously, HNO and NO differentiate pharmacologically and physiologically. The same may be true for potassium channel activation. It is widely accepted that NO is an EDHF, and that NO is capable of activating calcium-activated potassium channels (K+ ca) in the resistance vasculature, hyperpolarizing the VSM and causing relaxation.74 In contrast, investigators have shown conclusively that HNO

54 activates voltage-gated potassium channels (K+ v) in rat and mouse mesenteric arteries. -

55 The literature surrounding K+ v subtype expression is limited, but using a specific K+ v channel blocker, 4-AP, we were able to examine this phenomenon pharmacologically.

As expected, K+ v channel blockade decreased A Ch sensitivity in sham mesenteric arteries, but interestingly, we observed an almost complete abolishment of relaxation in vessels from the Angil mice. Given that these channels are specifically activated via

HNO, we believe that HNO-activation of K+ v channels in the mesenteric arteries to be dysfunctional during Angil hypertension. Also, these results were only observed in the resistance vasculature. We observed an endothelium-independent decrease in HNO­ mediated relaxation, via AS, only in the mesenteric arteries and not in the aorta 29

(unpublished observations), further demonstrating a role for HNO as an EDHF in resistance arteries. Moreover, CRCs to AS in the presence of the Kv + channel blocker elicited no effect in either group. We believe this suggests that AS mediates relaxation primarily in an ODQ-dependent manner, whereas Kv+ channels are activated by HNO in an endothelium-dependent manner, this is not a primary mechanism in AS-mediated relaxation.

The nonadrenergic/noncholinergic neuromodulator _calcitonin gene-related peptide (CGRP) is a potent vasodilator and has been linked to HNO-mediated vasorelaxation. It has also been postulated to be a primary mechanism for the positive ionotropy seen in cardiac tissue by HNO. 53 Intravenous instillation of AS or other HNO donors, such as IPA/NO, does not lead to increases in plasma cyclic guanosine

39 60 84 monophosphate (cGMP), but actually increases levels of CGRP. ' , Although CGRP can mediate various signaling pathways in VSM, such as increased NO and cyclic adenosine mono-phosphate ( cAMP), the use of CGRP antagonists does not prevent AS­ mediated vasodilation, suggesting that HNO may mediate cardiac muscle via CGRP, it is

39 53 unlikely that the vasodilatory effects ofHNO occur through CGRP. ' However, CGRP has been suggested as a biomarker for HNO activity in the vasculature, and we used

Western blot analysis to determine levels of CGRP in the mesenteric arteries of sham and

Angll-treated mice for this purpose. We found that protein expression of CGRP was significantly decreased in the mesentery of Angll-treated mice, suggesting a decrease in

HNO bioactivity, supporting our ex vivo data. 30

Various EDHF's have been discovered, characterized and studied during hypertension. Most have been met with contention from the scientific community; this is the same with the discovery of HNO. Many published papers demonstrate the complex pathways which mediate vascular smooth muscle relaxation, and any aberrance can have detrimental effects on vascular tone. We reveal that HNO-mediated vasorelaxation is decreased in mesenteric arteries from AngII hypertensive mice, which is supported with molecular data showing that CGRP, a biomarker for HNO activity, is similarly decreased.

Our data are the first to describe a role for HNO in AngII hypertension, and to suggest that K+ v channel activation may account for this dysfunction. 31

·so

'40 ---Bham '20 -+-1.\ngll

·oo

8] ---.....---.....------,..-----, Baseline Day1 Dsy5

Figure 1. Chronic collection of MAP in conscience mice; sham and Angil-treated mice. During control period, no difference in MAP was seen in either group. Within 5 days after mini-pump implantation, AngII treated mice exhibited increased MAPs. Data are represented as mean± SEM; n=3-6. *p<0.05, AngII vs. Sham 32

0 # 0 C :g::.t .. t C 0 25 0 :.:; 25 (ts .¾ X i (ts >< t!J 50 -a; 50 -a; 0:: "'" 0::: Sham E(+) ":i• 0~ -. ~ 75 75 • .. 0 Cl Sham E(-) ...... 100 ■ Sham Angll E(+) 0 Angll 100 0• Angil E(-) -9 ..s -7 -6 -5 -4 -9 ..s -7 -6 -5 -4 log (ACh[M]) log {Angeli's Salt[M))

Figure 2. Mesenteric arteries from Angil-hypertensive mice exhibit decreased ACh­ and AS-mediated relaxation responses. (a) Relaxation responses to ACh were performed in Phe (10 µM) contracted mesenteric arteries from sham surgery mice (Sham) and Angll-treated (Angll) mice. Relaxation responses were calculated relative to the contraction elicited by Phe. Data are represented as mean± SEM; n=4-9. tp<0.0001, EC50 values of Angll+HS vs. Sham. (b)

CRCs to the nitroxyl anion donor, AS, were performed in Phe (10 µM) contracted mesenteric arteries. HNO-mediated relaxation was determined in mesenteric arteries from intact, E(+), or denuded, E(-), Angll-treated [Angll E(+) and Angll E(-)] and sham

[Sham E(+) and Sham E(-)] mice. Relaxation responses were calculated relative to the contraction elicited by Phe. Data are represented as mean± SEM; n=S-10. #p<0.01,

ECso values of E(-) vs. E(+) in sham, tp<0.001, ECso values of E(-) vs. E(+) in Angll. 34

µM). Data are represented as mean ± SEM; n=S-9. *p<0.05, Rmax values of L­ cys+CPTIO vs. vehicle. 35

0 -•-~ .... l t C 0 25 '! i ' .... I ><(U 4) 50 0:: Sham 0~ 75 ■ Cl Sham/4-AP 100 Angil ()• Angll/4-AP -9 -8 -7 -6 -5 -4 log (ACh[Ml)

Figure 4. An inhibitor of voltage-gated potassium channels decreases relaxation in both sham and Angil-treated mice.

Concentration response curves to ACh were performed in ·Phe (10 µM) contracted first order mesenteric arteries from AngII (AngII) and sham (Sham) mice in the presence of the voltaged-gated potassium channel blocker, 4-AP (1 mM). Data are represented as mean± SEM; n=6-9. tp<0.001, Rmax values of 4-AP vs. vehicle; t0.0001, ECso values of 4-AP vs. vehicle 36

0 0 - -~-t~ + C C 0 0 :;:: 25 25 - '~Li 'i (G i )( ! )( .t -+ ('O cu Q) 50 Q) 50 0::: 0::: C~ 75 ■ Sham ~C 75 ■ Sham C Sham/4-AP C Sham/ODQ 100 Angll 100 Angil •0 Angll/4-AP •0 Anoll/ODQ i i i -9 -8 -7 -6• -5• -9 -8 -7 -6 -5 -4 log (Angeli's Salt[M]) log {Angeli's Salt[M])

(a) (b)

Figure 5. Relaxation responses to the nitroxvl anion donor, Angeli's Salt, are mediated through soluble guanylate cyclase.

Concentration response curves to AS were performed in Phe ( 10 µM) contracted first order mesenteric arteries from Angil (Angil) and sham (Sham) mice in- the presence of the (a) voltage-gated potassium channel blocker, 4-AP (100 mM) and the (b) soluble guanylate cyclase inhibitor, ODQ (1 µM). Data are represented as mean ± SEM; n=5-8. t0.0001, Rmax values of ODQ vs. vehicle in sham; t0.0001, ECso values of ODQ vs. vehicle in Angil+HS. 37

CGRP sGC ~, ~-actin ~-act in

Calcltonln Gene Related Peptide Guanylate Cyclase P1 1.25 * S' ~ 1.00 "' !? 100 ·1§ 0.75 '~~::1E nlli :> :> 80 _ E0.50 {! 0.25 <{ 0.00 !'!L---.-i-----.LI ------1-I Shom Angll+HS Sh~m An911+HS (a) (b)

Figure 6. Protein expression of calcitonin-gene related peptide is decreased and soluble guanylate cyclase B, increased in mesenteric arteries from Angil-treated mice.

Western blot analysis was performed on protein isolated from mesenteric arteries of

Sham and AngII - treated mice. Western blot data was densitized using the average of protein expression in each gel nonnalized to P-actin. Data from multiple gels are expressed a fold change (arbitrary units, AU) of Sham. Data are represented as mean ±

SEM; (a) n=4-5, #p<0.01; (b) n=4-5 . *p<0.05. 38

B. Aorta from angiotensin II hypertensive mice exhibit preserved nitroxyl anion

mediated relaxation responses.

ABSTRACT:

Hypertension is a disorder affecting millions worldwide, and is a leading cause of death and debilitation in the United States. It is widely accepted that during hypertension and other cardiovascular diseases the vasculature exhibits endothelial dysfunction; a deficit in the relaxatory ability of the vessel, attributed to a lack of nitric oxide (NO) bioavailability. Recently, the one electron redox variant of NO, nitroxyl anion (NO) has emerged as an endothelium-derived relaxing factor (EDRF) and a candidate for endothelium-derived hyperpolarizing factor (EDRF). NO- is thought to exist protonated

(HNO) in vivo, which would make this species more resistant to scavenging. However, no studies have investigated the role of this redox species during hypertension, and whether the vasculature loses the ability to relax to HNO. Thus, we hypothesize that aorta from angiotensin II (Angll)-hypertensive mice will exhibit a preserved relaxation response to Angeli's Salt, an HNO donor. Male C57Bl6 mice, aged 12-14 weeks were implanted with mini-osmotic pumps containing Angil (90ng/rriin, 14 days plus high salt chow) or sham surgery. Aorta were excised, cleaned and used to perform functional studies in a myograph. We found that aorta from Angil-hypertensive mice exhibited a significant endothelial dysfunction as demonstrated by a decrease in acetylcholine (ACh)­ mediated relaxation. However, vessels from hypertensive mice exhibited a preserved response to Angeli's Salt (AS), the HNO donor. To confirm that relaxation responses to 39

HNO were maintained, concentration response curves (CRCs) to ACh were performed in the presence of scavengers to both NO and HNO (carboxy-PTIO and L-cys, resp.). We found that ACh-mediated relaxation responses were significantly decreased in aorta from sham and almost completely abolished in aorta from AngII-treated mice. Vessels incubated with L-cys exhibited a modest decrease in ACh-mediated relaxations responses. These data demonstrate that aorta from AngII-treated hypertensive mice exhibit a preserved relaxation response to AS, an HNO donor, regardless of a significant endothelial dysfunction.

Introduction

Hypertension is a leading contributor to death and disability in the United States and globally. Although a variety of factors including, diabetes, smoking and other life style choices can be correlated with hypertension, there is still a subset of the population which becomes hypertensive without the presence of these risk factors. These patients develop what is known as essential hypertension and frequently do not respond consistently to commonly used anti-hypertensive medications.

Among other factors, it is widely accepted that vascular tone contributes to increased vascular resistance leading to systemic blood pressure. Many investigators have demonstrated, using hypertensive animal models, that increased blood pressure can alter the vascular contractility and relaxation through a variety of processes. One such pathway is through a decrease in nitric oxide (NO) production and/or bioavailability, as

NO is known to be a potent modulator of vascular tone. 56 NO is produced through the one electron reduction of L-arginine to citrulline and NO by nitric oxide synthase (NOS). 40

Since the discovery of an endothelium-derived relaxing factor (EDRF) and the determination of NO to be the EDRF, the vast majority of research has focused on this redox species of NO. Recent evidence suggests that a redox variant of NO, the nitroxyl anion (NO-), may mediate relaxation through potassium (K+) channel activation and

VSM hyperpolarization leading investigators to suggest that NO- is an endogenously­ derived EDHF. 25

Very little research exists on the other two redox variants of NO; the charged nitrosonium cation (NO+) and the one-electron product, nitroxyl anion (N0-).39' 56 It was thought that NO- would exist as an anion (pKa 4.7) at a physiological pH; however, this assumption was corrected in 2002, when investigators determined the actual pKa to be around 11.4 and that at physiological pH NO- exists as HNO.30-32 It was also determined that this conjugated weak acid, HNO, would be able to cross cellular membranes, leading

investigators to divert research to this understudied molecule.32 It is now known that the physiology, pharmacology and biochemistry of HNO are vastly different than that of

N0.21, 85

HNO and NO can both be produced through the conversion of L-arginine to NO by nitric oxide synthase (NOS), in the presence of required substrates; arginine, nADPH

and oxygen with the cofactors calmodulin and tetrahydrobiopterin (BH4) (for constitutive

NOS). 19' 37, 45 However, HNO is mainly produced when NOS is uncoupled or in the 437 69 absence of BH4_3 - , During these same conditions, the product of uncoupled NOS, N­ hydroxyl-L-arginine, can lead to HNO production. 70 During hypoxic conditions, NO can be generated via reduction of nitrites on metal sites and from conversion via xanthine

oxidase (XO). Interestingly, it was demonstrated that through a similar, yet aerobic 41

22 45 86 pathway that HNO can be produced. ' , Furthermore, HNO can be produced through the reduction of NO by mitochondrial cytochrome c (mCyC) and non-enzymatically from the decomposition of S-nitrosothiols. Other reports suggest HNO production to be a

87 88 41 56 71 product of the catalytic turnover of NOS. • , ,

Moreover, the mechanism by which HNO elicits its effects varies from NO.

Given the chemistry of HNO, the production and target site may be limited to the membrane, suggesting channels and other membrane-bound enzymes as possible

44 45 downstream targets. - This confirms reports of HNO S-nitrosylating the ryanodine receptor and myofilaments, causing both lusitropy and inotropy in cardiac tissue.4243 It has also been shown that HNO can activate adenosine-triphosphate (ATP)-sensitive potassium channels (KATP) in the coronary vasculature and voltage-gated potassium

54 56 channels (K+ v) in mesenteric arteries. - There is also evidence for both NO and HNO to activate soluble guanylate cyclase (sGC), and using in vitro techniques, cyclic

29 39 46 guanosine monophosphate (cGMP) accumulates in large conduit vessels only. • •

Infusion of HNO donors, such as Angeli's Salt, does not lead to an increase in systemic cGMP in contrast to NO donors, but to an increase in calcitonin gene-related peptide

(CGRP). CGRP is a potent vasodilator released from neurons and has been described as

39 53 a biomarker of HNO activity. • Although there is an increasing body of knowledge regarding HNO-mediated relaxation, no current studies have investigated the relative role that HNO plays during hypertension. Given the stability of HNO as compared to NO, the ability of HNO to maintain its ability to mediate vasorelaxation during hypertension may preclude HNO to be a possible therapeutic agent to exploit. We hypothesize that 42

HNO-mediated relaxation will be preserved in aorta from angiotensin II (Angil) hypertensive mice.

MATERIALS AND METHODS

Male C57bl/6 mice, weighing between 25-30 grams were obtained from Jackson

Laboratories (Bar Harbor, ME). Mice were maintained on a 12-hour light dark cycle, housed five per cage and allowed access to chow and water ad libitum. Isoflurane (10%) in oxygen was used for surgeries with carbon dioxide (CO2) for euthanasia. All procedures were performed in accordance with the Guiding Principles in the Care and

Use of Animals, approved by the Medical College of Georgia Committee on the Use of

Animals in Research and Education.

Blood Pressure Telemetry Studies

Mice were anesthetized using isofluorane and a DSI Data Transmitter (Data

Sciences International, St. Paul, MN) was implanted in the left carotid artery, routed and secured sub-scapularly. Animals were allowed to recover; systolic/diastolic pressures, heart rate and activity were collected for 18 hours per day/night. Data were analyzed using Power Lab (AD Instruments, Colorado Springs, CO). The blood pressure data from these mice are shared, as the mesenteric arteries and aorta were used in two separate papers.

Functional Studies

After euthanasia with CO2, the aorta was rapidly excised and bathed in ice-cold physiological salt solution (PSS) (NaCl 120 mM, KCl 4.7 mM, KH2PO4 1.18 mM,

NaHCO3 14.9 mM, dextrose 5.6 mM, CaChH2O, 0.06 mM EDTA). Increased 43 concentrations of EDTA were used in PSS buffer to aid in preventing the extracellular conversion of HNO to NO. Aorta were carefully isolated and mounted as ring preparations on two stainless steel pins in a myograph (Danish MyoTech, Aarhus,

Denmark). Vessels were maintained at 37° C and continuously aerate~ with 95% 0 2, 5%

CO2 and allowed to stabilize for at least 45 mins, at an optimal passive force of 5.0 mN.

After stabilization, tissues were contracted with KCl (120 mM) solution to determine the reactivity of the vascular smooth muscle cells. To determine the viability of the endothelium, contraction was stimulated via phenylephrine (Phe; 1 µM) followed by acetylcholine (A Ch; 10 µM). Vessels were then washed before performing concentration response curves (CRC) and after each CRC. CRCs to ACh or Angeli's Salt (AS,

Cayman Chemical, Ann Arbor, MI) in Phe-contracted vessels was performed in the presence of vehicle or the following inhibitors: carboxy-PTIO (CPTIO, nitric oxide scavenger), L-cystiene (L-cys, nitroxyl anion scavenger), 4-aminopyridine (4-AP, K+v channel blocker). CPTIO was obtained from Cayman Chemical and 4-AP obtained from

Tocris Bioscience, Ellisville, MO. All other chemicals and drugs were purchased from

Sigma Aldrich, St. Louis, MO. Force measurements were collected using Chart™

Software (ADI Instruments, Colorado Springs, CO) for PowerLab data acquisition systems (ADI Instruments).

Statistical Analysis

Agonist concentration-response curves were fitted using a nonlinear interactive fitting program (GraphPad Prism, Graph Pad Software Inc., San Diego CA), and values expressed as percent of maximal relaxation graphed against increasing molar 44 concentrations of agonist. Agonist potencies and maximum response are expressed as negative logarithm of the molar concentration of agonist producing 50% of the maximum response (ECso) and maximum effect elicited by the agonist (Rmax), respectively. Non­ linear regression analysis was used to determine ECso values, where Rmax was normalized to 100 percent for calculations. Data are expressed as mean±SEM (n), where n is the number of experiments performed. Statistical analysis of the concentration-response curves was performed by using the F test for comparisons of best-fit data between groups

(EC50 and Rmax), For CRCs where a best-fit analysis could not pe performed, absolute

Rmax values were averaged and Students' t-test performed for significance.

Drugs

Carboxy-PTIO was suspended in DMSO and Angeli's Salt was prepared using a

0.01M NaOH solution. All other stock solutions were prepared by using water. Stock solutions originally diluted in DMSO or ethanol was used with a final concentration of less than 0.003% v/v in the muscle bath; this concentration has been demonstrated to have no effect on vascular reactivity. Additionally, solutions containing vehicle levels of ethanol and DMSO were also used throughout the experimental protocol to control for non-specific effects.

RESULTS

Angiotensin II treated mice exhibited an increase in mean arterial pressure, as shown by telemetry.

Before drug treatment, MAP, HR and activity data were obtained during a control period of no less than 4 days in all mice. Mice treated with Angil for two weeks 45 exhibited a significant increase in MAP by day 3 (l 60±5mmHg, p<0.05). There were no observed changes in MAP, HR or activity for the sham surgery mice, whether given normal chow or high salt chow.

Aorta from Ang/I mice did not exhibit a decrease in AS-mediated relaxation responses despite a reduction in ACh-mediated relaxation reponses.

Aorta from Angll hypertensive mice exhibited a significant decrease in sensitivity to ACh as compared to sham aorta (EC5o -5.86±0.18 vs. -6.85±0.09, p<0.001). There was also a reduction in the maximal relaxation in Angll-treated aorta as compared to sham aorta (Rmax 37.00%±3.66 vs. 72.56%±2.46, p<0.001) (Figure 7). However, we observed no change in AS-mediated relaxation responses (Figures 8a and 8b) in Angll-treated mice as compared to vessels from sham mice. Similar results were obtained whether aortas were intact (E+, Figure 8a) or denuded (E-, Figure 8b ).

Aorta from AngII hypertensive mice are dependent upon nitric oxide for relaxation.

To further investigate the role of HNO in vasorelaxation, scavengers of NO and

HNO (CPTIO and L-cys, resp.) were used. Aortas were incubated with CPTIO and

CRCs performed to ACh. In the presence of CPTIO, vessels from sham mice exhibited a significant decrease in both sensitivity (EC50 -5.68 ± 0.34 vs. -6.85 ± 0.09, p<0.05) and maximal relaxation (Rmax 22.29% ± 5.48 vs. 72.56% ± 2.46, p<0.05) (Figure 9a) to ACh.

When aortas were incubated with CPTIO, there was almost a complete inhibition of relaxation to ACh (Rmax p<0.00l)(Figure 9b). As shown in Figure 4, aortas were incubated with the HNO scavenger, L-cys, which has been demonstrated as a mechanism 46

54 72 73 to differentiate between HNO and NO. · • In aorta from sham mice, a significant decrease in maximal relaxation was observed in vessels incubated with L-cys as compared to vehicle (Rmax 58.00% ± 3.43 vs. 72.56% ± 2.46, p<0.001) (Figure 10a).

However, there was no decrease in A.Ch-mediated relaxation in aorta from Angil­ hypertensive mice incubated with L-cys as compared to those without (Rmax 16.80% ±

3.37 vs. 36.99% ± 3.66)(Figure 10b). These data suggest a significant dependence upon

NO for vasorelaxation, and that NO bioavailability is decreased during Angil hypertension.

Aorta exhibit a decrease in ACh-mediated relaxation with voltage-gated potassium channel blockade.

The K+ v channel has been demonstrated to be specifically activated by HNO in rat and mouse mesenteric arteries. Given this, the role of K+ v channels in this model of hypertension was investigated. Aortas were incubated with 4-AP, which has been previously demonstrated to be a specific K+ v channel blocker. 25· 54 In Figure 11, aortas were incubated with the K+ v channel blocker or vehicle and CRCs to A Ch were performed. Vessels from sham animals exhibited a rightward shift in sensitivity to ACh

(EC50 vs. -6.85 ± 0.09, p<0.05), with no significant changes in maximal relaxation responses (Figure Sa). Aorta from Angil hypertensive mice trended towards a decrease in relaxation responses, but no significant changes were observed (Figure 11 b ). These data suggest that the K+ v channel may modulate a portion of endothelium-mediated relaxation. 47

AS-mediated relaxation is soluble guanylate cyclase dependent.

To assist in determining the mechanism involved in HNO-mediated relaxation,

CRC to AS were performed in aorta from sham and Angil-treated mice. Vessels were incubated with 4-AP, the Kv+ channel blocker, as shown in Figure 12a. No significant differences in sensitivity to AS were observed aorta from sham mice; however, vessels from Angil-treated mice exhibited a significant decrease in AS-mediated relaxation (-

5.19 ± 0.15 vs.-6.54 ± 0.16, p<0.0001). In vessels incubated with ODQ, the sGC inhibitor (Figure 12b), aorta from both sham (Rmax 27.0%) and Angil (Rmax 11.5%) - treated mice exhibited almost a complete inhibition of relaxation, suggesting that in the aorta, AS-mediated relaxation is largely dependent upon sGC for relaxation responses.

CRCs to AS were also performed in the presence of the NO and HNO scavengers,

CPTIO and L-cys, resp. We found that vessels from both sham (ECso -5.82 ± 0.08, p<0.001) and Angil-treated (EC5o -5.18 ± 0.14, p<0.001) mice, incubated with CPTIO exhibited a significantly decreased sensitivity to AS (Figure 13a). When similar curves were performed in the presence of L-cys, aorta from both sham (EC50 -5.89 ± 0.12, p<0.05) and Angil-treated (EC50 -5.53 ± 0.21, p<0.01) mice exhibited a decrease in AS­ mediated relaxation responses (Figure 13b).

DISCUSSION

The free radical species of nitrogen oxide, NO, is the most well known and well­ studied, in contrast to its redox congeners: NO- and NO+. As mentioned previously, studies involving NO- were not' pursued due to the fact that NO- was thought to exist as an anion in vivo; however, with new data it is known that NO- would actually be 48 protonated (HNO).30·33 As HNO, no channel or ion pore is needed for intracellular access, thus making HNO a possible mediator in cell signaling pathways.32 Studies have been performed using the HNO/NO- donor, AS, which decomposes to yield NO- and NO2 to determine the mechanism by which this species mediates relaxation in the vasculature.51 · 82· 89-90 Although literature shows a definite role for HNO-mediated relaxation, no studies to date have investigated the role which HNO plays in vessels from hypertensive animals. In this study, we set out to determine if HNO-mediated relaxation responses would be preserved in the aorta from AngII hypertensive mice, given that HNO is thought to exhibit decreased reactivity in the presence of reactive oxygen species

(ROS).

Previous studies have demonstrated that AS induces relaxation in vascular and non-vascular smooth muscle including: aorta, mesenteric artery, cornonary artery,

54 56 61 72 82 91 92 anococcygeus and gastric fundus. · · · • · - In the present study, we also found that the aorta from mice exhibited a similar half-maximal relaxation reponse to AS, 0.65

µM, when compared to other tissues (Figure 7a). What was interesting was our observation that relaxation responses to AS were maintained in aorta from AngII-treated mice, whether the vessels were intact or denuded (Figure 8). When CRCs to ACh were performed, aortas from hypertensive mice exhibited a significantly decreased sensitivity and maximal relaxation response, suggesting there is a profound endothelial dysfunction

(Figure 7). This study is first to demonstrate this maintenance of HNO-mediated vasorelaxation, and this phenomenon was in contrast to other findings from this laboratory showing that mesenteric arteries from AngII-hypertensive mice exhibit a 49 decrease in HNO-mediated relaxation. These data demonstrate regional differences in vascular reactivity of HNO during hypertension.

In responses mediated by ACh, we found that incubation with CPTIO (Figure 9) produced a marked decreased in relaxation, while incubation with L-cys (Figure 10) did not. These data indicate that endogenous production of NO, such as by ACh-mediated mechanisms, is scavenged in the presence of CPTIO. The presence of CPTIO in CRCs performed to ACh reduced the response in aorta from sham, confirming the role that NO plays in aortic vascular tone. Conversely, the almost complete abrogation of ACh­ mediated relaxation responses in aorta from Angil-treated mice also signifies the loss of

NO bioavailability for relaxation. L-Cys incubation produced a modest, though significant decrease in ACh-mediated relaxation in vessels from sham, but not from

Angil-treated mice revealing little reliance upon HNO for relaxation. When compared to our data using the HNO donor, AS, we believe that these data may also suggest that there may be different mechanisms for relaxation when HNO is produced endogenously.

In order to investigate mechanisms for relaxation, we also performed CRCs to AS and ACh in the presence of the sGC inhibitor, ODQ, and the Kv+ channel blocker, 4-AP.

When CRCs to ACh were performed, aorta from both mice exhibited a decrease in ACh sensitivity, while those from Angil-treated mice a decrease in maximal relaxation.

However, when CRCs to AS were performed, only vessels from Angil-treated mice exhibited a rightward shift in the relaxation response. In preliminary results, we found that Kv+ channel blockade has a profound effect on endothelium-mediated relaxation, which, given that resistance vessels rely heavily upon EDHF for relaxation, is expected. 50

Our current data demonstrates a role for Kv+ channels in modulating aortic vascular tone and possible Kv+ channel in hypertension. To further investigate this role, CRCs were performed to AS in the presence of the Kv+ channel antagonist, with surprising results.

Only vessels from Angil hypertensive mice exhibited a decrease in AS-mediated relaxation. When similar CRCs were performed in the presence of ODQ, blocking sGC, we observed a complete inhibition of AS-mediated relaxation. We believe these data reveal that AS-mediated relaxation is completely dependent upon sGC for relaxation, and again suggesting that endogenously produced HNO may differ in mechanism from exogenously applied HNO.

All CRCs were performed in the presence of the copper (Cu(ll)+) chelator, EDTA, in order to prevent or decrease the extracellular conversion of HNO to NO, as shown by

55 72 82 83 other investigators. ' , - Using NO-sensing probes, Favaloro and colleagues demonstrated that Cu+ present in the PSS used during experiments will oxidize NO- to

NO. 55 Although the use of EDTA will limit the extracellular conversion of NO- to NO, there is limited evidence regarding the intracellular conversion between these two molecules. In our experiments, we found that both L-cys and CPTIO attenuated AS­ mediated relaxation responses in aorta from sham and Angil-treated mice. Although this is the first study to investigate HNO-mediated relaxation in aorta from Angil hypertensive mice, there are other studies to demonstrate that both scavengers reduce AS­ mediated relaxation responses. In a study performed by Ellis and co-workers, they found that AS mediated relaxations differently in the aorta than in the non-vascular tissue, anococcygeus muscle. They also found that CPTIO itself, oxidizes NO- to NO, albeit · less than 2% of the total amount of AS added. 82 In our present study, we also observed a 51

decrease in AS-mediated relaxation responses in the presence of CPTIO. Interestingly,

the scavenging effect of CPTIO was greater in aorta from the AngII-treated mice. Given

that a possible intracellular oxidative effect may be inter-converting these two molecules, the fact that this effect was increased in aorta from hypertensive mice gives us insight into different possible oxidative environments of these vessels. Other laboratories have shown that there are increases in superoxide dismutase (SOD) activity during AngII hypertension, which may also explain our findings. 93 If vascular disease and hypertension lead to increases in expression and/or activity of other metal containing enzymes such as xanthine oxidase (XO), which is highly likely given the oxidative environment during these conditions, this would also contribute to changes in HNO/NO interconversion. Additionally, with increasing concentrations of AS (> 10 µM), there have been shown to be increases in NO, along with nitrite production. 55 Other

investigators have ruled out a possibility of nitrites to be mediating the vast majority of effects seen by AS, given that nitrites mediate vascular relaxation at concentrations much

55 72 higher than what is present with AS decomposition. , We also observed a further attenuation of AS-mediated relaxation responses in aorta from AngII-treated mice when incubated with L-cys as compared to aorta from sham mice incubated with L-cys.

Investigators have used single doses of AS and found that millimolar levels of L-cys

significantly decrease the maximal relaxation responses. L-cys, in lower concentrations, is suggested to either block NO-mediated vasorelaxation. With- low concentrations (<

300 µM) of L-cys, Furchgott and Feelisch et al. found that these concentrations blocked the response to NO and to EDRF, which they surmised to be through the formation of

15 72 73 94 96 superoxide during the auto-oxidation of L-cys. ' - , - Moreover, data suggest that 52 via S-nitrosothiol production, L-cys may potentiate NO-mediated responses in the

15 72 73 vasculature when used at high concentrations (> lmM). • - In our preparation, we did not find a potentiation of ACh-mediated relaxation responses in the presence of L-cys.

This again calls into question the oxidative environment of the aorta, as there were observed differences in AS-mediated relaxation responses in the presence of L-cys in aorta from sham and Angil-treated mice.

Overall, our data are the first to demonstrate that aorta from Angil-treated hypertensive mice, while displaying endothelial dysfunction also exhibit a preservation of

HNO-mediated relaxation responses. We also demonstrate that there.may be a difference in the oxidative environment of the aorta, as compared to other vascular and non-vascular smooth muscle tissues. 53

0 C: 0 i 25 >< (II 50 a::~ ¢~ 75 ■ Sham • Angll 100

-9 -8 -7 -6 -5 -4 log (ACh[M))

Figure 7. Aorta from Angil-hypertensive mice exhibit endothelial dysfunction. Concentration response curves to ACh were performed in Phe (1 µM) contracted aorta.

ACh-mediated relaxation responses were assessed in aorta from Angil-treated (Angil) and sham (Sham) mice. Relaxation responses were calculated relative to the maximal contraction elicited by Phe. Data are represented as mean± SEM; n=12-18. tp<0.001,

ECso and Rmax values of Angll+HS vs. Sham. 54

0 C: t~:fl-~~ 0 0 C 25 0 i ~~ i 25 ~ >< 'i. 50 ~ ,,b -;; 60 o:= ct= f. [J i,~ Sham E(+} 75 Sham E(-) ~¢ 75 • 0 Angil EM '-1;(·0 • Angll E(+) 'l!l 100 100 I -9 .a -7 -6 -5 -4 -9' -8 -7' -6' -5' -4' log (Angeli's Salt[M]) log {Angeli's Salt[M])

(a) (b)

Figure 8. Angeli's Salt-mediated relaxation responses were preserved in intact and denuded aorta from Angil hypertensive mice.

Concentration response curves to the nitroxyl anion donor, Angeli's Salt, were performed in Phe (1 µM) contracted aorta. Relaxation responses to nitroxyl anion were assessed in intact aorta from Angll-treated (Angll,) and sham (Sham) mice. Nitroxyl anion-mediated relaxation was determined in aorta from intact, E( +), (a) Angll-treated (Angll) and sham

(Sham) mice and denuded, E- (b). Relaxation responses were calculated relative to the maximal contraction elicited by Phe. Data are represented as mean± SEM; n=12-16. 55

.c...g_§! 0 0 .... i .... C C 0 0 !,i-, 25 25 i)( i * (U X (0 -a; 50 a> 50 * ~ Angll "''#- Sham "0 75 • 75 C• Sham/CPTIO 0 Angll/CPTIO 100 100

-9 -8 -7 ·6 .5 -4 -9 -8 -7 -6 -5 -4 log (ACh[M]) log (ACh[M])

(a) (b)

Figure 9. Scavenging nitric oxide decreases ACh-mediated relaxation responses in aorta from Angll-treated and sham mice. Concentration response curves to ACh were performed in Phe (1 µM) contracted aorta from sham (Sham,)(a) and Angll (Angll)(b) in the presence of the nitric oxide scavenger, carboxy-PTIO (200 µM). Data are represented as mean± SEM; n=7-9. *p<0.05, ECso and Rmax values of CPTIO vs. vehicle in sham; tp<0.001, Rmax values of CPTIO vs. vehicle in Angll. 56-

0 0 C C 0 0 i 25 i 25 >< X ~ cu 50 50 a> ~ ~ 0~ Angil ~ 75 ■ Sham 75 • a Sham/L-cys 0 Angll/L-cys 100 100

-9 -8 -7 -6 -5 -4 -9 -8 -7 -6 -5 -4 log (ACh[M]) log (ACh[M])

(a) (b) Figure 10. Nitroxyl anion does not primarily mediate endothelium dependent vasorelaxation in aorta from both sham and Angll-treated mice. Concentration response curves to ACh were performed in Phe (1 µM) contracted first aorta from sham (Sham)(a) and Angll (Angll)(b) in the presence of the nitroxyl anion scavenger, L-cys (3 mM). Data are represented as mean± SEM; n=9-l 1. tp<0.001,

Rmax values of L-cys vs. vehicle in sham. 57

0 0 4'1-JJ C: C: 0 0 25 ;:; 25 '~* i co ~, >< co>< ('IS 50 Q) 50 ~ ~'~-f ~ ~0 Angil ~0 75 Sham 75 • C• Sham/4-AP 0 Angll/4-AP 100 100

-9 -8 -7 -6 -5 -4 -9 -8 -7 -6 -5 -4 log (ACh(M]) log (ACh[M])

(a) (b)

Figure 11. Voltage-gated potassium· channel blockade decreases relaxation in aorta.

Concentration response curves to ACh were performed in Phe (1 µM) contracted aorta from sham (Sham)(a) and Angil (Angll)(b) in the presence of the voltaged-gated potassium channel blocker, 4-AP (1 mM). Data are represented as mean± SEM; n=6-9.

*p<0.05, EC5o values of 4-AP vs. vehicle in sham. 58

..ZS-~-~ :t: 0 ---,~ C 0 =i 25 ,, 'l >< \ (,s ' -a; 50 ~ Sham <>~ ■ ', Sham 75 'f. a Sham/4-AP ' Sham/ODQ 100 Angll Angil 0• Angll/4-AP Anell/ODQ I i I i -9 -8 -7 -0 -5 -8 -7 -6 -5' -4 log (Angeli's Salt[M]) log (Angeli's Salt[M]}

(a) (b)

Figure 12. Relaxation responses to the nitroxyl anion donor, Angeli's Salt, are mediated through soluble guanylate cyclase.

Concentration response curves to AS were performed in Phe ( 1 µM) contracted aorta from Angll (Angll) and sham (Sham) in the presence of the (a) voltage-gated potassium channel blocker, 4-AP (1 mM) and the (b) soluble guanylate cyclase inhibitor, ODQ (1

µM). Data are represented as mean± SEM; n=S-8. tp<0.0001, EC5o values of 4-AP vs. vehicle in Angll, tp<0.0001, Rmax values of ODQ vs. vehicle in sham and Angil. 59

:j: .Q-.0-Q 0 0 0 i§ '" tl .... ~ # C: C: 0 .... ~ ' 0 '~"~ i 25 t' ~ i 25 >< >< C'G ~ \ (V '~ 'I -; 50 'fi ~, q) 50 * 0:: 'Q'-l,J~ ~ Sham ~, '1:S ~ Sham Q 75 • "0 75 ■ C Sham/CPTIO □ Sham/L-cys '~ 100 Angil 100 Angll 0• Angll/CPTIO 0• An~I 1/L-c~s I i • .9 -s -1 -6 -5 .9' -8 .7 -$' -5' -4 ,og (Angeli's Salt[M]) log (Angeli's Salt[M])

(a) (b)

Figure 13. Relaxation responses to the nitroxyl anion donor, Angeli's Salt, are reduced with nitric oxide and nitroxyl anion scavenging.

Concentration response curves to AS were performed in Phe ( 1 µM) contracted first aorta from AngII (AngII) and sham (Sham) in the presence of the (a) NO scavenger,

CPTIO (200 µM) and the (b) HNO scavenger, L-cys (3 mM). Data are represented as mean± SEM; n=S-8. *p<0.05, ECso values ofL-cys vs.vehicle in sham; #p<0.01 ECso values of L-cys vs.vehicle in AngII; tp<0.001, ECso values of CPTIO vs. vehicle in sham, ip<0.0001, ECso values ODQ vs. vehicle in AngII. 60

C: Angeli's Salt, a nitroxyl anion donor, reverses endothelin-1 mediated vascular

dysfunction in aorta.

ABSTRACT:

Although nitroglycerin (Gtn) 1s a clinically prescribed treatment for cardiovascular patients due to its vasodilatory actions, it is widely accepted to induce tolerance when given chronically. A proposed mechanism for this process is the superoxide (0-2)-oxidative stress hypothesis, which suggests that Gtn increases NADPH oxidase superoxide (0-2) production. Nitric oxide (NO) exists in three different redox states; uncharged NO, the oxidized state (NO+), and the reduced state, nitroxyl anion

(NO-). NO- has newly become an emerging candidate in vascular regulation given the pharmacological and biochemical differences that exist between NO- and NO·. Only recently was evidence shown that NO- is produced ex vivo, in an endothelium-dependent manner (A Ch) and is a potent vasodilator of both conduit and small resistance vessels. In addition, since NO- exits protonated (HNO), it is more resistant to scavenging proposing

HNO to be of particular interest in pathophysiological conditions where high levels of reactive oxygen species (ROS) exist, such as hypertension. We hypothesize that treatment with Gtn will exacerbate ET-1 vascular dysfunction via an increase in ROS, while treatment with AS will not via an increase in ROS. Aorta from C57Bl6 mice, aged

12 weeks, were isolated and incubated in DMEM overnight, per standard cell culture protocol. Vessels were divided into four groups: vehicle, ET-1 [1 µM], ET-1 + Gtn [1

µM], ET-1 + AS [1 µM]. Concentration response curves (CRCs) to acetylcholine.(ACh) and phenylephrine (Phe) were performed. We found that aorta incubated with ET-1 61 exhibited a significantly decreased relaxation response to ACh and an increase in Phe­ mediated contraction. Vessels incubated with Gtn still exhibited a significant decrease in

ACh-mediated relaxation and Phe-mediated contraction while aorta incubated with AS exhibited a complete reversal in vascular dysfunction. When vessels were incubated overnight with the superoxide dismutase (SOD) mimetic, tempol, we observed an improvement in endothelial function, as demonstrated by an increase in ACh-mediated relaxation. In addition, using primary aortic cell culture, we found that incubated with

ET-1 increased superoxide (02-) production, as visualized by DHE staining. This increase in 02- was maintained with Gtn treatment, while decreased when co-incubated with AS. These data suggest that aorta incubated with the HNO donor, AS, can reverse

ET-1 mediated vascular dysfunction, which may be through a decrease or prevention of

ROS generation. We propose that HNO may be vasoprotective and that HNO donors may be studied as a possible therapeutic option where other organic nitrates are contraindicative.

Introduction

Nitroglycerin (glyceryl nitrate, Gtn), was proposed by Murrell in 1879 as a

97 98 treatment for heart failure. - Since then, Gtn continues to be administered to alleviate various cardiovascular conditions, including angina, heart failure and other acute hypertensive crises. 62 While this, and other organic nitrovasodilators, may improve cardiac output by decreasing preload and afterload on the heart, as well as inducing vasodilation, these compounds are also of limited clinical efficacy. Organic nitrates are known to induce tolerance when given over time, which has been demonstrated clinically 62

as well as in various animal models. Additionally, when conditions of increased ROS are

present, they are also thought to have a diminished vasoprotective effect, and even induce

vascular wall damage.

Of late, there has been considerable research regarding the one electron reduced

39 56 congener of nitric oxide (NO), nitroxyl anion (N0-). • It was thought that NO- would exist as an anion (pKa 4. 7) at a physiological pH; however, this assumption was corrected

in 2002, when investigators determined the actual pKa to be around 11.4 and that at

30 32 physiological pH NO- exists as HNO. - It was also determined that this conjugated weak acid, HNO, would be able to cross cellular membranes, leading investigators to

divert research to this understudied molecule. 32 Although the mechanisms of relaxation

and the question of endogenous sources are still being investigated, it is widely accepted that physiology, pharmacology and biochemistry of HNO.are vastly different than that of

NO.21, 85

This phenomenon can be especially seen when using the HNO/NO- donor,

Angeli's Salt (AS). Various studies have used AS to study HNO-mediated vasorelaxation, and one of the most attractive effects of AS is the lack of tolerance

induced. In ex vivo functional studies performed using rat aorta, Irvine and coworkers

demonstrated that use of the HNO-donor AS, did not induce tolerance, as compared to

Gtn. Additionaly, they revealed that AS did not even exhibit a cross tolerance to Gtn. 61

A recent study has demonstrated a similar phenomenon in vivo, which demonstrated that

61 63 chronic infusion of AS did not lead to tolerance or endothelial dysfunction. • The

exact mechanisms of Gtn-induced tolerance have not been fully elucidated; however,

several proposed models have been recommended. These mechanisms may overlap, but 63

are suggested to include: reduced biotransformation of the organic nitrates to NO, neurohormonal activation, desensitization of soluble guanylate cyclase (sGC), increased

61 66 98 101 phosphodieasterase lAl activity and increased production of ROS. • • - In fact, there is a growing body of literature showing that administration of organic nitrates not

only induces tolerance, but promotes vascular wall damage.

It has also been demonstrated that HNO is less reactive, as compared to NO, which may lend HNO superior stability against reactions with ROS or during conditions

29 39 58 69 81 102 of high oxidative stress, such as hypertension. ' • • ' • Common mediators of

hypertension induced ROS generation are angiotensin II and endothelin-1 (ET-1 ), which

have been implicated in the promotion of oxidative stress and vascular wall damage.

Given this, we evaluated whether use of the HNO/NO- donor, AS, would alleviate the

Gtn induced vascular dysfunction induced by ET-1. We hypothesize that treatment with

Gtn will exacerbate ET-1 vascular dysfunction via an increase in ROS, while treatment

with AS will not.

MATERIALS AND METHODS

Male C57bl/6 mice, weighing between 25-30 grams were obtained from Jackson

Laborotories (Bar Harbor, ME). Mice were maintained on a 12-hour light dark cycle,

housed five per cage and allowed access to chow and water ad libitum. Isoflurane (10%)

in oxygen was used for surgeries with carbon dioxide (CO2) for euthanasia. All

procedures were performed in accordance with the Guiding Principles in the Care and

Use of Animals, approved by the Medical College of Georgia Committee on the Use of

Animals in Research and Education. 64

Primary Cell Culture

Aorta from mice were carefully excised and cleaned in sterile Dubelco' s Modified

Eagle's Medium (high glucose, DMEM), supplemented with 1% penicillin/streptomycin and 30% fetal bovine serum (FBS). Per previous protocols and experimental procedures used in this laboratory, endothelium was removed and vascular smooth muscle cell

(VSMC) culture obtained via explant technique. 103 VSMC primary cell culture was maintained in DMEM (low glucose), supplemented with 1% penicillin/streptomycin and

10% FBS under normal cell culture conditions (37°C, 95% 0 2, 5% CO2). Cells were grown to confluency, passed using trypsin and used within 4-6 passages.

Aortic Ring Incubation

After euthanasia with CO2, the aorta was rapidly excised and bathed in ice-cold physiological salt solution (PSS) (NaCl 120 mM, KCl 4.7 mM, KH2P04 1.18 mM,

NaHC03 14.9 mM, dextrose 5.6 mM, CaChH20, 0.06 mM EDTA). Increased concentrations of EDTA were used in PSS buffer to aid in preventing the extracellular conversion of HNO to NO. Aorta were carefully isolated and incubated in DMEM (low glucose) supplemented with 1% penicillin/streptomycin. Vessels were incubated overnight in incubator per standard cell culture technique and divided into the following groups: vehicle, endothelin-1 (ET-1, O.lµM or 1 µM, Pheonix Pharmaceuticals), endothelin-1 plus the nitroxyl anion donor, Angeli's Salt (AS, 1 µM, Cayman Chemical) or endothelin-1 plus nitroglycerin (Gtn, 1 µM, American Regent, Shirley, NY). Vessels were incubated overnight in either vehicle or ET-1 for 20-22 hrs. During the last hour of incubation, vessels were given two treatments of either AS or Gtn ( one treatment at one hour prior and second 30 minutes prior to functional experiment). 65

Functional Studies

Aorta were mounted as ring preparations on two stainless steel pins m a myograph (Danish MyoTech, Aarhus, Denmark). Vessels were maintained at 37° C and continuously aerated with 95% 02, 5% CO2 and allowed to stabilize for at least 45 mins, at an optimal passive force of 5.0 mN. After stabilization, tissues were contracted with

KCl (120 mM) solution to determine the reactivity of the vascular smooth muscle cells.

To determine the viability of the endothelium, contraction was stimulated via phenylephrine (Phe; 1 µM) followed by acetylcholine (ACh; 10 µM). Vessels were then washed before performing concentration response curves (CRC) and after each CRC.

CRCs to ACh in Phe (10 µM)-contracted vessels were performed in the presence of vehicle or the following: tempol (02- scavenger), ebselen (ONOO- scavenger), Fe-TPPS

(ONOO- degradation catalyst). All other chemicals and drugs were purchased from

Sigma Aldrich, St. Louis, MO. Force measurements were collected using Chart™

Software (ADI Instruments, Colorado Springs, CO) for PowerLab data acquisition systems (ADI Instruments).

Dihydroethidium Staining

VSMC primary cell culture was used in determining reactive oxygen species

(ROS) and superoxide (02) generation with dihydroethidium (DHE, Invitrogen,

California), which oxidizes in the presence of ROS, generating fluorescence. VSMC were grown to confluence, and then passaged to 6-well plates. Once VSMC reached 80% confluency, experiments were performed using: vehicle, ET-1 (0.1 µM), ET-1 plus AS

(10 µM) and ET-1 plus Gtn (10 µM). VSMC were treated with ET-1 for 2 minutes, in 66 the presence of vehicle, AS or Gtn administered simultaneously. After washing 3 times in phosphate buffered saline (PBS, lX) VSMC were incubated with DHE (2 µM) for 20 minutes. VSMCs were washed with PBS and then fluorescence visualized using (Zeiss,

Carl Zeiss Microlmaging, Thomwood, NY).

Statistical Analysis

Agonist concentration-response curves were fitted using a nonlinear interactive fitting program (GraphPad Prism, Graph Pad Software Inc., San Diego CA), and values expressed as percent of maximal relaxation graphed against increasing molar concentrations of agonist. Agonist potencies and maximum response are expressed as negative logarithm of the molar concentration of agonist producing 50% of the maximum response (EC5o) and maximum effect elicited by the agonist (Rmax), respectively. Non­ linear regression analysis was used to determine EC50 values, where Rmax was normalized to 100 percent for calculations. Data are expressed as mean±SEM (n), where n is the number of experiments performed. Statistical analysis of the concentration-response curves was performed by using the F test for comparisons of best-fit data between groups

(ECso and Rmax),

Drugs

Angeli's Salt was prepared using a 0.0lM NaOH solution. All other stock solutions were prepared by using water. Stock solutions originally diluted in DMSO or ethanol was used with a final concentration of less than 0.003% v/v in the muscle bath; this concentration has been demonstrated to have no effect on vascular reactivity.

Additionally, solutions containing vehicle levels of ethanol and DMSO were also used throughout the experimental protocol. 67

RESULTS

Aortic rings incubated with endothelin-1 exhibit vascular dysfunction.

To establish our model of vascular dysfunction, aortas were treated with ET-1

(0.1 µM and 1 µM) overnight and vascular reactivity assessed by performing CRCs to

Phe and ACh. In Figure 14, aortic rings treated with ET-1 (0.1 µM), did not exhibit an increased contractility to Phe (Rmax 105.1% ± 2.06 vs. 110.0% ± 2.56) (Figure 14a), nor a decreased endothelium-mediated relaxation response to A Ch (EC5o -6.84 ± 0.11 vs. -6.95

± 0.12) (Figure 14b). When the concentration of ET-1 was increased to [1 µM], a significant vascular dysfunction was observed. ET-1 treated aortic rings exhibited a significantly enhanced Phe-mediated contraction, as compared to those incubated with vehicle alone (Rmax 128.3% ± 3.5 vs. 115.5% ± 3.26, p<0.01) (Figure 14c). CRCs were then performed to ACh, and ET-1 treated aortic rings exhibited a significantly decreased

ACh-mediated relaxation response, indicative of endothelial dysfunction (Rmax 58.33% ±

3.2 vs. 81.68% ± 2.6, p<0.001) (Figure 14d). For the remainder of the experiments, the higher ET-1 treatment (1 µM) dose was used.

Incubation with Angeli's Salt, the HNO donor, reversed the increase in Phe-mediated contraction induced by ET-1.

To determine if Gtn would exacerbate ET-1 mediated vascular dysfunction, CRCs were performed in aortic rings incubated with ET-1 overnight and co-treated with Gtn (1

µM) twice in the last hour. As shown in Figure 15a, vessels incubated with ET-1 exhibited a significantly increased Phe-mediated contraction, which was not significantly 68 altered with Gtn co-treatment (Rmax I2I .7% ± 2.37 vs. I28.3% ± 3.46). In contrast, when

ET-I vessels were treated with AS, there was a significant decrease in Phe-mediated contraction, as compared to those incubated with ET-I alone (Rmax I I 7.2% ± 3.58 vs.

I28.3% ± 3.46, p<0.05), which was not different than that of vehicle (Figure I5b).

These data reveal that AS may be able to reverse ET-I-mediated vascular dysfunction.

Incubation with the HNO donor, Angeli's Salt, reverses ET-I-mediated endothelial dysfunction.

To determine if Gtn treatment would further exacerbate ET- I-mediated endothelial dysfunction, CRCs were performed to ACh. Vessels were treated with ET- I or vehicle overnight, and then given Gtn twice in the last hour. Aortic rings incubated with ET-I exhibited a marked endothelial dysfunction (Rmax 58.33% ± 3.2 vs. 81.68% ±

2.6, p<0.001) (Figure 16). When these vessels were co-incubated with Gtn, there was an improvement in endothelial dysfunction (Rmax 71.3% ± 2.6 vs. 58.3% ± 3.2, p<0.05)

(Figure 16a), yet relaxation responses remained significantly decreased as compared to vehicle. When these vessels were treated with AS, a complete reversal of endothelial dysfunction was observed (Rmax 83.2% ± 2.9 vs. 58.3% ± 3.2, p<0.0001) (Figure 16b), suggesting AS may be capable of improving ET-1 induced endothelial dysfunction.

Endothelin-1 treatment increases superoxide generation, which is decreased when co­ treated with Angeli's Salt. 69

Using a primary vascular smooth muscle (VSMC) culture, experiments were performed to further investigate the mechanism of AS vasoprotection. VSMCs obtained from murine thoracic aorta were maintained in cell culture for 4-6 passages before use, which has been demonstrated to be an acceptable time period during which VSMC maintain their phenotype in culture. 104 VSMCs were treated with: vehicle, ET-1 (0.1 µM,

2 min), ET-1 plus AS (10 µM, 2 min) or ET-1 plus Gtn (10 µM, 2 min). After treatment,

DHE staining was performed to visualize 0 2 • generation. As shown in Figure 17b, ET-1 treatment increased 02· generation, as compared to basal levels. When VSMCs were incubated with Gtn (Figure 17d), no changes in DHE staining were apparent. However, when VSMCs were incubated with AS (Figure 17c), a decrease in DHE staining was observed, suggesting that AS may be preventing or reducing 02· generation in smooth muscle cells.

Superoxide scavenging improved vascular reactivity in ET-1 treated aortic rings.

To determine a mechanism for the improvement in vascular function seen with

AS incubation, vessels were treated with the ONOO- decomposition catalyst, FeTPPS (10

µM) or the superoxide dismutase (SOD) mimetic, tempol (lmM).

To determine if 0 2• and NO were reacting to form ONOO-, resulting in decreased vascular function, vessels were incubated with FeTPPS (FT) for 30min prior to CRCs.

Interestingly, ET-1 treated aorta incubated with FeTPPS exhibited a marked decrease in

ACh sensitivity (EC50 -6.19 ± 0.09) and maximal relaxation response (Rmax 52.0% ± 2.4)

(Figure 18a). In addition, incubation with FeTPPS in ET-1 +Gtn (Figure 18b) resulted in 70 no significant changes while FeTPPS incubation in ET-1 +AS (Figure 18c) reduced ACh­ mediated relaxation responses (Rmax 56.l % ± 3.85 vs. 83.2% ± 2.85, p<0.0001).

In order to determine if the prevention or reduction of 0 2• would induce similar effects as incubation with the HNO donor, AS, tempol was used. These aorta were incubated with tempol concurrently with ET-1; vessels were then incubated overnight before co-treatment with either AS or dtn. Vessels incubated with Gtn and co-incubated with tempol exhibited no changes in ACh-mediated relaxation responses as compared to

ET-1 +Gtn (Figure 19a) (Rmax 71.2% ± 3.6 vs. 71.3% ± 2.6). When ET-1 treated vessels were similarly incubated with tempol and AS, no additional effects of tempol were observed (Figure 19b) (Rmax 75.2% ± 2.4 vs. 83.2% ± 2.9). However, ET-1 treated vessels (Figure 19c) incubated with tempol alone exhibited a significant improvement in the maximal ACh-mediated relaxation responses (Rmax 82.6% ± 2.6 vs. 58.3% ± 3.2, p<0.0001), with no changes in the sensitivity to ACh. Taken together, these data suggest that removal or reduction of either ONOO- or 0 2• does not have a similar vasoprotective effect as incubated with the HNO donor, AS.

DISCUSSION

The use of nitrosovasodilators, such as Gtn, for the treatment of cardiovascular conditions is well established and has been used for decades. Although this class of vasodilators mediate rapid and capable vasodilation, they are limited clinically due to the development of tolerance and reported increases in ROS generation. 105 The redox variant of NO, nitroxyl anion, has recently become the subject of increased investigation.

HNO/NO- has been demonstrated to induce vasodilatation in an array of vascular beds 71

comparable to NO. Furthermore, researchers have clearly established that AS does not

induce vascular tolerance, using ex vivo preparations, and more recently, in an in vivo

61 63 chronic model. ' However, there are no data demonstrating whether the use of HNO,

a more stable gas, would also decrease the vascular wall damage seen with NO donors.

In this study we proposed that the NO donor, Gtn, would exacerbate ET-1 mediated

vascular dysfunction and that this would be alleviated by using the HNO donor, AS.

The role of NO as an EDRF in the vasculature has been intensively studied and

has been clearly established as an important modulator of vascular tone and relaxation.

NO elicits VSM relaxation through soluble guanylate cyclase (sGC) and protein kinase G.

(PKG) activation, as well as direct activation of BKca channels. In addition to its

vasodilatory properties, NO can also reduce platelet aggregation, increase angiogenesis

and has potent anti-inflammatory capabilities. The redox variant of NO, NO-, once

determined to exist protonated at a physiological pH (HNO), has been the subject of

intense investigation and emerged as an EDRF and as a candidate of EDHF (need ref for

25 54 56 106 pH here ). ' , , Like NO, HNO/NO- was revealed to have a similar anti-aggregating

effect on platelets, and is protective during ischemia-reperfusion events however, this is

59 64 where the similarities end. •

The mechanism of HNO-mediated vasorelaxation is still.poorly understood,. but

likely involves voltage-gated potassium channel (Kv +) activation in resistance vessels,

adenosine triphosphate (ATP)-sensitive potassium (KATP) activation in coronary arteries

54 56 and through calcitonin-gene related peptite (CGRP) in the capacitance vasculature. -

Although sensitive to sGC inhibition by ODQ, infusion of an HNO donor does not lead to 72 an increase in systemic cyclic guanosine monophosphate (cGMP) but an increase in

CGRP (reference here). There are still more questions than answers regarding the role of endogenous HNO in the vasculature and probable physiological source.

There is an abundance of literature demonstrating the limited clinical efficacy of organic nitrates, such as Gtn, and the phenomenon of nitrate tolerance and cross-tolerance is well-documented, but yet the mechanism still not clearly defined. Several hypotheses exist including; diminished bioconversion of Gtn, neurohumoral adaptations, and vascular oxidative stress. Although it is suspected that the process is multifactorial, the

100 107 latter process has emerged as a principal factor. • It has been demonstrated that treatment with Gtn increased ROS generation using platelets, endothelial and smooth

100 108 muscle cells as well as in animal models. • Other studies have demonstrated a similar mechanism in patients. Schulz and colleagues revealed that patients treated with

Gtn for 24 hours exhibited endothelial dysfunction and increased ROS production in

105 109 addition to nitrate tolerance. • In this manuscript, we report that Gtn does not exacerbate ET-1 mediate vascular dysfunction, but slightly improves Phe-mediated contractile and ACh-mediated relaxation responses (Figures 14a and 15a). In this study, the model used was to simulate multiple doses, or a chronic Gtn model. However, there are questions as to whether more doses or an increase in incubation time would alter the results we observed.

Various enzymatic sources of ROS have been proposed to induce Gtn tolerance, including; xanthine oxidase (XO), nicotinamide adenine diphosphate (NADPH) oxidase and mitochondrial ROS production. However, there is a definite link between the 73

NADPH oxidase family of enzyme and their contribution to tolerance induced ROS

99 110 111 generation and more specifically vascular 0 2• production. ' • In 2006, Otto et al. demonstrated that apocynin, an inhibitor ofNADPH oxidase, conferred protection against nitrate tolerance, and a year later, Fukatsu and colleagues demonstrated that 02 · production was decreased in Gtn-treated endothelial cells when NADPH oxidase was

99 112 inhibited. '

In this study, we used an ET-1 induced model of endothelial dysfunction. ET-1 has been consistently demonstrated to induce hypertension, endothelial dysfunction and

ROS. Our data demonstrate that overnight incubation ofmurine aorta with ET-1, induces an increased Phe-mediated contraction and a decrease in ACh-mediated relaxation

(Figure 14). The underlying ROS production was a perfect modelto determine if co­ incubation with Gtn would exacerbate the vascular dysfunction present (Figures 15a and

16a). As mentioned previously, we found that Gtn improved vascular function, although relaxation and contraction responses remained significantly different to vehicle treated aorta. Although surprised by these data, it is difficult to compare this model with studies using in vivo and chronic Gtn treatment.

The most significant finding of the current study was that two treatments of AS, the HNO donor, during the last hour of ET-1 incubation could completely reverse the vascular dysfunction. It has been shown that AS has positive effects when infused chronically, and does not lead to tolerance or cross-tolerance to other NO-donors.

However, there are no studies demonstrating whether HNO treatment can reduce ROS generation. 74

As shown in Figure 17, we observed an increase in ROS production, or 02· generation with ET-1 treatment, which is a known contributor to vascular dysfunction and is present during hypertension and other cardiovascular disease. We hypothesized that giving Gtn, or NO as a free radical gas, may lead to increased ROS production and peroxynitrate formation (ONOO}

In order to determine if the increased ET-1 induced ROS generation is reduced with AS treatment; DHE staining was utilized in order to visualize ROS/ 02· production in a primary VSM cell culture model (Figure 17). VSM cells which were incubated with

ET-1 exhibited a greater DHE staining. When the VSM cells were co-incubated with AS there was a decrease in DHE staining which did not occur with Gtn co-incubation. These data suggest that AS may be reducing the levels of VSM cell ROS/ 02· production.

To confirm these data, functional studies were performed as before, but with vessels co-incubated with the superoxide dismutase mimetic, tempol (Figure 19).

Investigators have shown that tempol has a variety of beneficial vascular effects. We did not· observe any differences in ET-1 treated vessels incubated with AS or Gtn and co­ incubated with tempol. In contrast, when ET-1 treated vessels were co-incubated with tempol, a significant improvement in maximal relaxation response were observed. When taken in conjunction with the experiments performed in cell culture, these data suggest that AS may be exerting vasoprotective effects through a decrease in 02· production, given that tempol treatment did improve the relaxation response. However, tempol treatment did not confer a similar level of vasoprotection as treatment with AS salt alone.

In addition, vessels which were co-treated with tempol did not exhibit improved 75 relaxation in the ET-l+Gtn vessels. Taken together, these data suggest that 02· production alone is not entirely responsible for the effect of AS treatement.

To further investigate our hypothesis and determine if the increase in 02· production may be reacting to produce increased levels of ONOO-, CRCs were performed in vessels incubated with the ONOO- degradation catalyst, FeTPPS (Figure

18). However, when ET-1 +Gtn treated vessels were incubated with FeTPPS, we found that there was no increase in ACh-mediated relaxation. Furthermore, ET-1 treated aorta

incubated with FeTPPS alone worsened the endothelial dysfunction. Although confusing, there is emerging data suggesting a regulatory role for ONOO-, and so the quantity of ONOO- necessary to maintain proper vascular function may have been compromised in this study. From our data, we can conclude that ET-1 induced ONOO­ production may not be a primary cause for the vascular dysfunction present in this model.

Overall, our data demonstrate that AS, the HNO donor can confer vasoprotection in ET-1 mediated vascular dysfunction. We believe that this mechanism may be through

VSM cell ROS/ 02· production. Additional mechanisms contributing to this phenomenon include. The use of HNO donors may prove to be a beneficial therapeutic alternative to organic nitrates in the near future. 76

150 0

C 125 C: 0 25 ~ 100 i ftS >

# 0 125 _,i_J C: C: 0 ,Q 100 25 ..ca 0 cu >< .. 75 ca 50 C: ai -0 0:: -1..J 0 50 ?fl. :::Ii! 75 0 25 ■ Vehicle ET-1 Vehicle 0 ■ 100 • • ET-1 -9 -8 -7 -6 -5 -4 -9 -8 -7 -6 -5 -4 log (Phe[M]) log (ACh[M])

(c) (d)

Figure 14. Aorta incubated with endothelin-1 exhibit vascular dysfunction.

Concentration response curves to Phe performed in aorta incubated with [0.1 µM] (a) and

[ 1 µM] (c) ET-1. Concentration response curves to A Ch were performed in aorta incubated with [0.1 µM] (b) and [1 µM] (d) ET-1 and pre-contracted with Phe (10 µM).

Relaxation responses were calculated relative to the maximal contraction elicited by Phe.

Data are represented as mean ± SEM; n=4-9. #p<0.01, lp<0.0001, ET-1. vs. vehicle. 77

# 125 125

100 100 ;:;5 ;:;5 (,) (,) co 75 (0 ....'"" .... 75 C: '""C: 0 50 0 0 0 50 ';I. 25 ';I. 25 • Vehicle • Vehicle 0 ET-1 0 ET-1 •CJ ET-1+Gtn 0• ET-1+AS -9 -8 -7 -6 -5 -4 -9 -8 -7 -6 -5 -4 log (Phe(M]) log (Phe[M])

(a) (b)

Figure 15. Incubation with AS reverses ET-1 induced vascular dysfunction in mouse aorta. Concentration response curves to Phe performed in intact aorta incubated with either endothelin-1 (ET-1), ET-1 plus nitroglycerin (ET-l+Gtn [1 µM]) (2a) or Endothelin-1 plus Angeli's Salt (ET-l+AS [1 µM]) (2b). Data represented as mean±SEM, n=9-ll,

#p<0.01, ET-1 vs. vehicle,# p<0.01, ET-1 +AS vs. ET-1. 78

0 0 C C O 25 0 i i 25 >< >< CV ~ 50 0,) 50 Vehicle ~ ~ ■ '#. ■ Vehicle 75 • ET-1 JJt 75 • ET-1 □ ET-1+Gtn 0 ET-1+AS 100 100

-9 -8 -7 -6 -5 -4 -9 -8 -7 -6 -5 -4 log (ACh[M]) log (ACh[M])

(a) (b)

Figure 16. Incubation with the HNO donor, Angeli's Salt, reverses ET-1 mediated endothelial dysfunction.

Concentration response curves to ACh were performed in Phe (10 µM) aorta. (a).

Relaxation responses to ACh were assessed in intact aorta incubated with vehicle, ET-1

[1 µM] or ET-1 +Gtn [1 µM]. (b) Relaxation responses to ACh were assessed in intact aorta incubated with vehicle, ET-1 [ 1 µM] or ET-1 + AS [ 1 µM]. Relaxation responses were calculated relative to the maximal contraction elicited by Phe. Data are represented as mean± SEM; n=S-10. *p<0.05, ET+Gtn vs. vehicle, tp<0.001, ET-1 vs. vehicle. 79

a. vehicle b. ET-I

C. ET-l+AS d. ET-l+Gtn

Figure 17. Endothelin 1 treatment increases superoxide generation, which is decreased when co-incubated with Angeli's Salt.

Dihydroethidium (DHE) staining for superoxide production in a primary cell culture of aortic vascular smooth muscle cells (VSMC) Cells were treated with vehicle (a), endothelin-1 (ET-I [0.1 µM]) (b), ET-I plus Angeli' s Salt (ET-l+AS [10 µM]) (c) or ET­

I plus nitroglycerin (ET-I +Gtn [ 10 µM]) for 2 minutes, and then probed for superoxide production using DHE. 80

0 0 :j: C C 0 25 0 z .;- 26 cu \ ell )( ; cu 50 "i 50 "i Ct! Ct! -~- ~0 Vehicle 0~ 75 ■ 75 ■ Vehicle ET-1 · ET-1+Gtn • • ET-1 +Gtn+FeTPPS 100 Cl ET+FeTPPS 100 CJ

.9 -8 .7 -6 -5 -4 -9 -8 .7 -6 .5 -4 log (ACh[M]) log {ACh[M]) (a) (b)

0 C 0 i 25 co)< G) 50 a: ..I.l 0~ 75 Vehicle 0• ET-1+AS 100 o ET-1 +AS+FeTPPS

-9 -8 -7 -6 -5 -4 log (ACh[M]) (c)

Figure 18. Incubation with the ONoo· degradation catalyst, FeTPPS, reduces ACh­ mediated relaxation responses.

Concentration response curves to ACh were performed in Phe (10 µM) contracted aorta.

(a). Relaxation responses to ACh were assessed in intact aorta incubated with vehicle,

ET-1 [1 µM] or ET-1 +FeTPPS [10 µM]. (b). Relaxation responses to ACh were assessed in intact aorta incubated with vehicle, ET-1 [1 µM], ET-l+Gtn [1 µM] or ET-

1+Gtn+FeTPPS [10 µM]. 5c. Relaxation responses to ACh were assessed in intact aorta incubated with vehicle, ET-1 [1 µM], ET-l+AS [1 µM] or ET-l+AS+FeTPPS [10 µM].

Relaxation responses were calculated relative to the maximal contraction elicited by Phe.

Data are represented as mean± SEM; n=4-10. tp<0.0001, ET-1 +FeTPPS vs. ET-1,

tp<0.0001, ET-l+AS+FeTPPS vs. ET-1+-AS. 81

0 0 C C .S? 25 ;;0 25 ..C'IS C'IS >< >< .!!! 50 .!!! 50 ~ ~ ET-1+Gtn '#. '#. ■ 75 • ET-1+AS 75 a ET-1+Gtn+T 0 ET-1+AS+T 100 100

-9 -8 -7 -6 -5 -4 -9 -8 -7 -6 -5 -4 log (ACh[M]) log (ACh[M])

(a) (b)

C O 25

1 50 ~ '?ft 75 ■ Vehicle 100 • Et-1- a E-1+T -9 -8 -7 -6 -5 -4 log (ACh[M])

(c)

Figure 19. Incubation with tempol does not improve ACh-mediated relaxation responses in Gtn and ET-1 treated aorta.

Concentration response curves to ACh were performed in Phe (10 µM) contracted aorta incubated overnight with endothelin-1. (a). Relaxation responses to ACh were assessed in intact aorta incubated with the HNO donor, AS [1 µM], and with or without tempo! [1 mM] (a), with the NO donor, Gtn [1 µM], and with or without tempo! (b) and with or without tempo! (c ). Relaxation responses were calculated relative to the maximal contraction elicited by Phe. Data are represented as mean± SEM; n=5-10. tp<0.0001,

Et-1 +T vs. vehicle. 82

III. UNPUBLISHED DATA

Abstract

HNO/NO- has emerged as an important modulator of vascular relaxation in conduit and resistance arteries. The exact mechanisms accounting for the relaxation responses seen in mesenteric and aortic vascular smooth muscle (VSM) is still under investigation. It is postulated that HNO/NO- leads to soluble guanylate cyclase (sGC) activation, S-nitrosylation (SNO) events and voltage-gated potassium (K+ v) channel activation. To determine whether the HNO/NO- donor, Angeli's Salt (AS), can induce

K+ v channel activation and induce in vitro relaxation responses, whole-cell patch clamp and vascular reactivity studies were performed. We hypothesized that AS would produce

K+ channel activation and that concentration response curves (CRCs) to AS would produce vasodilation in mesenteric arteries with pharmacological blockade of other K+ channels necessary for relaxation. Male C57Bl6 mice, aged 12-14 weeks were killed and aorta and mesenteric arteries removed. A primary vascular smooth muscle cell (VSMC) culture was obtained using the explant method, and used for whole-cell patch clamp experiments. First order mesenteric arteries were isolated and used on a wire myograph for functiqnal studies. Using the whole-cell patch clamp technique, we observed that the application of AS induced increased K+ flux, which was abolished with the addition of the K+ v channel inhibitor, 4-aminopyridine (4AP). In addition, functional experiments revealed that AS was capable of producing vasodilation in the presence of inhibitors which selectively block SK, IK+ ca and BK+ ca calcium channels. These data demonstrate 83 that AS can induce K+ v activation, and that this activation can lead to vasorelaxation in mesenteric arteries.

Methods

Vascular smooth muscle cell isolation.

Aorta were dissected and cleaned from male C57 /Bl6 mice, aged 12-14 weeks.

Primary aortic VSMC culture was obtained and maintained through first passage. Cells were removed from dish, using trypsin, and re-suspended in dissociation buffer ( 110 mM

NaCl, 5 mM KCl, 0.16 mM CaCh, 10 mM HEPES, 0.5 mM NaH2PO4, 10 mM NaHCO3,

0.5 mM KH3PO4, 10 mM glucose, 0.49 mM EDTA and 10 mM taurine). For patch­ clamp experiments, several drops of the cell suspension were placed in a microscope chamber which contained the recording solution (140 mM NaCl, 5 mM KCL, 2 mM

MgCh, 2 mM CaCh, 20 mM HEPES and 20 mM glucose, pH 7.4). 113

Whole-cell currents were measured from metabolically intact aortic VSMC using the amphotericin perforated-patch technique. 113-116

Perforated patch-clamp experiments.

Perforated patch-clamp experiments were performed using pipettes fabricated from glass capillaries (7052, Corning, Lowell, MA) with a resistance of 3 n or less. In order to measure K+ current, the tips of the pipettes were filled with a solution containing

90 mM KCH3SO3, 40 mM KCl, 5 mM MgCh and 20 mM HEPES (pH 7.4) to approximate normal intracellular [K+] and [Cr]. The remainder of the pipettes was back­ filled with a similar solution, and 200 m/mL amphotericin B ( diluted by DMSO). Cells 84 were studied only if the voltage-drop across the series resistance was reduced to :S 5 m V within 10-20 min after forming a gigohm seal. Voltage clamp and voltage pulse generation was controlled with an Axopatch 200 A patch clamp amplifier (Axon

Instruments, Downington, PA) and data were analyzed.with pCLAMP 6.0.3 (Axon

Instruments, Inc.). Voltage-activated currents were filtered at 2 kHz and digitized at 10 kHz, and capacitative and leakage currents were substracted digitally. All drugs were diluted into fresh bath solution and perfused into a 0.5 mL recording chamber (Warner

Instruments Corp., Hamden, CT).

Functional experiments

Vascular reactivity studies were performed, as described earlier in first order mesenteric arteries.

Drugs

Angeli's Salt ([10 µL], AS, Caymen Chemical) was dissolved in 0.01 M NaOH.

Iberiotoxin ([500 nM], IbTx, Tocris) and 4-aminopyridine ([1 mM], 4-AP, Tocris) were dissolved in distilled water. All drugs were applied directly to the recording chamber.

All other reagents and equipment needed for the experiment were generously provided by Dr. Richard White, Medical College of Georgia.

Results

Angeli's Salt induces voltage-gated potassium channel activation in aortic vascular smooth muscle cells.

Whole-cell patch clamp experiments were performed in aortic VSMCs obtained from primary cell culture. In Figure 20, outward K+ currents were obtained in aortic

VSMC at baseline. After generation of currents from the voltage range of -40 mV to +50 85 mV, the HNO/NO- donor, AS, was directly applied to the recording chamber. The addition of AS induced an increase in outward K+ current at low voltages (Figure 20), and altered the current profile from that of BK+ ca channel to K+ v currents (Figure 21 ). In

Figure 20c, the addition of 4AP, the selective K+ v channel inhibitor abolished K+ currents induced by AS. After a washout period, partial K+ flux was recovered in the cell, demonstrating that AS induces and outward K+ channel current.

Angeli's Salt induced relaxation responses in mesenteric arteries via voltage-gated potassium channel activation.

CRCs to AS were performed in mesenteric arteries to determine whether the

HNO/NO- can produce relaxation responses in the presence of other K+ channel inhibitors. As shown in Figure 22, AS induces relaxation responses in the mesenteric arteries with an EC5o of approximately -6.3 M. When CRCs were performed to charybdotoxin (ChTx), iberiotoxin (lbTx), and apamin (AP), AS still induced a relaxation response the responses were significantly shifted rightward. In addition, when these inhibitors were used together, a similar effect of AS was observed. 86

Control AS AS+ 4AP Washout

(a) (b) (c) (d)

Figure 20. Angeli's Salt, the HNO donor, induces outward potassium current in aortic smooth muscle cells.

Potassium currents recorded from aortic VSMC, using amphotericin B, over entire range of current (-40 to +50 mV). Control recordings obtained from VSMC (a) over range of currents generated. Angeli's Salt (b)(AS, [10 µM]) was applied to recording chamber, producing an increase in outward potassium (K+) current. Addition of the voltage-gated potassium channel (K+ v) inhibitor, 4AP [1 mM], abolished outward K+ current ( c ). In panel (d), drug was washed out and partial recovery of K+ current observed. 87

eoo

400 >"s c,: -;e,.,. 2.t:lO• ~ -=-e 0 ~

~ =;.. --200 ~i ::s .400 0 r 50 too 150 200 I Time (seq

Figure 21. Angeli's Salt induces changes in current kinetics.

Isolated and enlarged currents of potassium (Kl at +30 mV during control, Angeli's Salt

(AS, [10 µM]) administration and after wash-out period demonstrating a change in kinetics from calcium activated potassium (BK+ ca) channels to voltage-gated potassium

(K+ v) channels. 88

0 C 0 i 25 > 50 0:: • Vehicle ,:,~ ChTx 75 •0 lbTx CJ ChTx+lbTx 100 ♦ ChbTx+lbTx+Ap • i i t ..9 ' -8' ..7 -6 -5 -4' Angeli's Salt (log[M]) Figure 22. Concentration response curves to Angeli's Salt in mesenteric arteries induces relaxation through voltage-gated potassium channel activation.

Concentration response curves (CRC) to Angeli's Salt (AS) were performed in phenylephrine (Phe [10 µM]) contracted first order mesenteric arteries. CRCs were performed in the presence of potassium channel inhibitors charybdotoxin (ChTx [100

µM]), iberiotoxin (IbTx, [100 nM]), apamin (AP, [1 µM]), ChTx+lbTx and

ChTx+IbTx+AP. 89

IV. DISCUSSION

The role of NO has been intensively studied since the discovery that NO was an

EDRF; however, there is limited knowledge regarding HNO in the vasculature. As mentioned previously, until recently HNO was simply regarded as a pharmacological phenomenon. This was until a paper published demonstrating that the pKa of Nff/HNO was greater than 11, which hinted at a possible endogenous role of HNO in the vasculature.33 Since that discovery HNO has emerged as an EDRF. Although limited in quantity, sufficient data exists that HNO is a vasodilator in various vascular beds; from rodent to human.

In the aorta, seminal studies by Fukuto and colleagues revealed that a clinically used drug, cyanamide, elicited vasodilation, which was not catalase dependent and thus

47 117 mediated by HNO. ' From there, it was determined that HNO not only produced

25 53 vasodilation in conductance arteries, but in coronary arteries and resistance vessels. ' -

56 60 54, • Even though there is an increasing body of knowledge, there have been no studies investigating the role of this redox species in the vasculature of hypertensive animals. As such, three specific aims were formed. The first was to determine the contribution of

HNO in the aorta and mesenteric arteries from hypertensive mice. The second was to delve into possible mechanisms for HNO-mediated relaxation. The last was to discern a possible role for HNO donors in conferring vasoprotection during an ET-1 induced model of vascular dysfunction.

The chemistry of HNO suggests that this gas is probably more stable and would

39 40 be less reactive with ROS. - Given the large amounts of ROS produced in the 90 vasculature during CVD, especially hypertension, the question of whether HNO­ mediated vasodilation is preserved during hypertension remained at large. 69 During conditions of oxidative stress and increased ROS, it is known that NO bioavailability is significantly decreased. This study demonstrates that the aorta from Angll-hypertensive mice exhibit a preserved AS-mediated relaxation response even though ACh-mediated relaxation responses are significantly decreased. In order to determine whether endogenously produced HNO is similarly affected in the hypertensive vasculature, CRCs to ACh were performed in the presence of scavengers for NO and HNO (C-PTIO and L­ eys, resp.). (See Figure 25 for schematic of HNO- and endothelium-mediated relaxation.) It was determined that aorta from normotensive mice exhibit a profound dependence upon NO for relaxation, as evidenced by a profound decrease in ACh­ mediated relaxation responses in the presence of the NO scavenger, C-PTIO. In aorta from Angll-hypertensive mice incubated with C-PTIO vasorelaxation was completely inhibited, suggesting that during hypertension NO bioavailability is significantly reduced.

In contrast, when CRCs to ACh were performed in the presence of L-cys only a slight decrease in maximal relaxation was observed in aorta from both normotensive and AngII­ hypertensive mice. It seems as though HNO does not play a primary role in the vasorelaxation of aorta.

When similar experiments were performed in first order mesenteric arteries an opposite effect was revealed. Mesenteric arteries from AngII-hypertensive mice exhibited a significant, endothelium-independent decrease in AS-mediated relaxation.

These data were confirmed using L-cys and CRCs to ACh. Mesenteric arteries from both 91 sham and hypertensive animals exhibited a significantly decreased endothelium­ dependent relaxation response when HNO was scavenged.

Even though these data are the first to determine the role of HNO in vessels from hypertensive mice, these data were not surprising. In the past year, HNO has been conclusively shown to induce dilations, which seem to be dependent upon potassium

39 55 channel activation. ' The conductance vasculature is more dependent upon NO and the resistance vasculature more dependent upon EDHF-type mechanisms, only supporting this current set of studies. HNO 'can' produce vasorelaxation, such as through the donor AS, in both vessel types, however, HNO seems to be more physiologically relevant in the small resistance vessels.

The exact mechanisms of HNO-mediated relaxation in vivo is still under debate, but work by Irvine et al. and Kemp-Harper et al. have revealed convincing evidence that

54 55 HNO induces potassium channel activation, specifically the Kv + channel. - There is limited data and so whole-cell patch clamp studies were performed with the help of Dr.

Richard White. Using aortic VSMCs from a primary cell culture, whole cell recordings were obtained. In these experiments, we observed that AS induced an increase in potassium flux, which was completely abolished when 4-AP, a specific Kv+ channel antagonist, was added to the preparation. Interestingly, the kinetics of the channels were also altered. Normal BKca channels were active in the VSMC previous to AS administration, which was shown by the slope and plateau of the curve. After AS administration, the profile of the curves was altered, revealing a quick change in potential and a tapering off of the curve. This is indicative of Kv+ channel activation. In functional studies using mesenteric arteries, we found that Kv + channel blockade 92

decreased ACh-mediated relaxation, yet almost completely inhibited relaxation in vessels

from Angil-hypertensive mice. Given that HNO has been demonstrated to activate Kv+

channels, we believe that HNO-mediated relaxation is decreased in mesenteric arteries

from Angil-treated mice via altered Kv+ channel activation (Figure 25). This process,

. coupled with a decrease in NO bioavailability, may contribute to the pathogenesis of

hypertension. Although these data support this conclusion, further studies will be needed

to conclusively demonstrate this event. While there is increasing confirmation of a role

for HNO as an endogenous mediator of vascular relaxation, new techniques and tools will

need to be uncovered in order to fully understand its role and production.

In addition, the regional variability demonstrated in these data; preservation of

HNO-mediated relaxation in aorta as compared to mesenteric arteries further suggests an

EDHF-like mechanism as shown here and by others. Vasorelaxation to ACh was

markedly reduced in the presence of the NO scavenger, CPTIO, in aorta, but not in the

mesenteric artery. Previously, investigators have shown an increasing reliance upon non­

NO vasodilators as the diameter of the vessel decreases. Furthermore, evidence

demonstrated by whole cell patch clamp reveals that AS produces a robust K+ flux, which

can be abolished with the use of the Kv+ channel antagonist. Using isolated vessel

preparations, these data were functionally confirmed using the Kv + channel antagonist in

CRCs to ACh. Although further investigation is needed, these data suggest that HNO

may be inducing K+ flux via the Kv+ channels, predominantly in resistance vessels.

The methods to detect HNO-mediated relaxation and/or by-products of HNO

bioavailability and function are extremely limited. Thus, CORP levels were assessed in

resistance vessels as a biomarker for HNO bioactivity. As mentioned previously, levels 93 of CGRP are increased with AS infusion, and investigators suggest that this can be used a differentiating marker between HNO and NO. We observed a decrease in CGRP levels in mesenteric arteries from the Angll-treated mice, which corroborates our functional data suggesting a decrease in HNO bioactivity or availability. However, further data are needed to fully understand the role which CGRP plays in HNO-mediated relaxation, and the accuracy of CGRP as a biomarker.

As mentioned previously, HNO 1s thought to be resistant to scavenging, especially with regards to 0 2-. Most, if not all, investigators agree that HNO-mediated vasodilation via a donor, such as AS, produces relaxation responses which are equal in magnitude and consistent with those stimulated by other nitrovasodilators. In contrast to other nitrovasodilators, such as Gtn, AS does not induce tolerance to itself or cross­

61 63 66 98 109 tolerance to other organic nitrates. • • • • In addition, research suggests that chronic use of Gtn can induce vascular wall damage, increased ROS generation and

66 105 107 109 118 endothelial dysfunction. • • • • Hence, we hypothesized that AS would not exacerbate endothelial dysfunction and ROS production in comparison to Gtn. A model of ET-1 vascular dysfunction was chosen, due to the fact that ET-1 stimulates production of high levels of ROS. At the time this study was proposed, very little research existed regarding overnight ET-1 incubation of murine aorta and the level of resulting vascular dysfunction. Accordingly, two doses of ET-1 were chosen for incubation. We found that only [1 µM] ET-1 led to a prominent decrease in ACh-mediated relaxation and an increase in Phe-mediated contraction. For the remainder of the study, that concentration was used. In addition, most studies looking at Gtn tolerance use a model of repeat and/or

66 100 105 109 119 chronic exposure. • , • • For that reason, the protocol of two doses within the 94 last hour of ET-1 treatment was developed as an ex vivo protocol for repetitive dosage.

CRCs were performed in ET-1 treated aorta, either given the AS or Gtn in the last hour.

Our results were surprising. We observed a complete reversal of vascular dysfunction in vessels incubated with AS. Although there was an unanticipated improvement in vessels incubated with Gtn, the improvement was modest and the ACh-mediated relaxation and

Phe-mediated contractile responses still remained significantly different from that of vehicle treated aorta. These data are the first to demonstrate a vasoprotective role for the

HNO donor, AS.

Current literature suggests that increased ROS production and ONOO- generation is a primary mechanism for Gtn tolerance. However, there were no improvements in vascular reactivity in ET-1 +Gtn treated aorta when incubated with the ONOO- scavenger,

FeTPPS. Moreover, only modest increases in relaxation responses were seen in vessels incubated with the SOD mimetic, tempol. These data suggest that Gtn tolerance· may be only partially due to ROS production. Also, the fact that two doses of AS within the last hour of a high ET-1 doses was able to reverse vascular dysfunction suggests that AS may be altering other pathways within the VSMCs leading to increase relaxation. Other experiments will be needed to further elucidate possible mechanisms. 95

VSM Relaxation VSM Relaxation

(a) (b)

Figure 23. Schematic of nitroxyl anion mediated relaxation via Angeli's Salt (a) and endothelium-mediated relaxation via acetylcholine. 96

HNO/No-

+.....___I Vascular Relaxation

t Peripheral Resistance

.. ! Hypertension

Figure 24. Schematic for increased vascular resistance and hypertension.

A decrease in HNO/NO- bioactivity, coupled with a decrease in NO bioavailability leads to a decrease in the ability of the vasculature to relax. This process contributes to increase peripheral vascular resistance, contributing to hypertension. 97

V. SUMMARY

Even though research of HNO/NO- has been increasing, there is still a deficit in current knowledge regarding the role of HNO-mediated relaxation. In addition, prior to these studies no data existed regarding HNO and hypertension existed.

From these studies we can conclude that AS-mediated relaxation responses are preserved in aorta from Angll-hypertensive mice, while mesenteric arteries from Angll­ treated mice.are not, despite a profound decrease in ACh-mediated relaxation. Consistent with other findings, the aorta relies heavily upon NO for relaxation and during Angll hypertension, NO bioavailability seems to be significantly decreased. In contrast, mesenteric arteries exhibit more dependence upon HNO for relaxation, which seems to be lost during Angll hypertension. By using whole-cell patch clamp studies, we believe that HNO, as released by AS, can preferentially activate Kv+ channels. Our functional data suggest that in mesenteric arteries HNO-mediate relaxation is lost via Kv+ channels, and this process may be contributing to the compromised endothelial function in Angll hypertension.

Moreover, these data demonstrate a possible difference in endogenous HNO­ mediated relaxation as compared to exogenous application, such as by the HNO donor

AS. In this study we also observed that AS was particularly sGC-dependent, which is similar to other investigators in the field. In contrast, AS-mediated relaxation was found to be only partially dependent upon Kv+ channel activation, whether in aorta or mesenteric arteries .. These results were surprising as Kv + channels seemed to contribute significantly to endothelium-mediated relaxation responses in the mesenteric arteries. It 98 should be noted that different laboratories have observed a variation in AS-mediated relaxation responses, which may be attributable to buffer concentrations (including chelating metal concentrations), oxygen levels, species and vascular bed differences and changes within the oxidative environment.

Treatment for hypertension is an underlying reason for this research, which led us to investigate possible therapeutic options for HNO donors, such as AS. We observed that treatment with AS completely reversed ET-1 mediated vascular dysfunction. These results were surprising, and give compelling evidence for HNO as a vasoprotective molecule. However, our investigations into the mechanism left many questions.

Although we found that AS seemed to decrease ROS/O2 • generation in an aortic VSM primary cell culture model, we were not able to replicate the effects of AS by using tempol or FeTPPS. These data suggest that AS may be mediating its effects through

ROS-independent mechanisms and given that HNO is highly thiophillic, S-NO is a possibility.39

In conclusion, these studies contribute to the evidence that HNO is an EDRF, inducing vasodilation in the aorta and mesenteric arteries. We also believe that HNO may be mediating vasorelaxation in an EDHF-like manner, due to the activation of the

25 55 Kv+ channels seen by us, and others. ' Our data suggest that HNO is an important mediator in vasorelaxation and that HNO-mediated relaxation is lost, possibly due to Kv + channels, in mesenteric arteries from Angll-hypertensive mice. In addition, the use of

HNO donors, such as AS may prove to be a useful alternative to organic nitrates during

CVD conditions such as angina or as a vasoprotective molecule. 99

VI. REFERENCES

1. Wynne BM, Chiao CW, Webb RC. Vascular Smooth Muscle Cell Signaling Mechanisms for Contraction to Angiotensin II and Endothelin-1. J Am Soc Hypertens. Mar-Apr 2009;3(2):84-95. 2. Michael SK, Surks HK, Wang Y, Zhu Y, Blanton R, M. J. High blood pressure arising from a defect in vascular function. Proc Natl Acad Sci -U S A. 2008; 105 :6702-6707. 3. Feletou MaV, P. M. Endothelial dysfunction: a multifaceted disorder. Am J Physiol Heart Circ Physiol. 2006;291:H985-Hl002. 4. Behrendt DG, P. Endothlial function: from vascular biology to clinical applications. Am J Cardiol. 2002;90:40L-48L. 5. Vanhoutte PM TE. Endothelium-dependent contractions: when a good guy turns bad! J Physiol. 2008;586:5295-5304. 6. Vanhoutte PM, Shimokawa H, Tang EH, Feletou M. Endothelial dysfunction and vascular disease. Acta Physiol (Oxf). Jun 2009; 196(2): 193-222. 7. Lyons D. Impairment and restoration of nitric oxide-dependent vasodilation in cardiovascular disease. Int J Cardiol. 1997;62:S101-Sl09. 8. Vanhoutte PM. Endothelial dysfunction and atherosclerosis. Eur Heart J Nov 1997;18 Suppl E:E19-29. 9. Joseph L. Izzo Jr. DAS, Henry R. Black, ed Hypertension Primer. 4 ed: Wolters Kluwer; 2008. 10. Nunes KP, Rigsby CS, Webb RC. RhoA/Rho-kinase and vascular diseases: what is the link? Cell Mo/ Life Sci. Nov 2010;67(22):3823-3836. 11. Hilgers RH, Webb RC. Molecular aspects of arterial smooth muscle contraction: focus on Rho. Exp Biol Med (Maywood). Dec 2005;230(11):829-835. 12. Furchgott RFaZ, J. V. The obligatory role of endothelial cells in the relaxation of arterial smooth muscle by acetylcholine. Nature. 1980;288:373-376. 13. Serban DN, Nilius, B., Vanhoutte, P. M. The endothelial sage: the past, the present, the future. Eur J Physiol. 2010;459:787-792. 14. Ignarro LJ, Byrns, R. E., Buga, R. M., Chaudhuri, G. & Wood, K. S. . Pharmacological evidence that endothelium-derived relaxing factor is nitric oxide: use of pyrogallol and superoxide dismutase to study endothelium-dependent and nitric oxide-elicited vascular smooth muscle relaxation. J Pharmacol Exp Ther. 1988;244:181-189. 15. Feelisch M, Te Poel, M., Zamora, R., Deussen, A. and Moncada,_ S. Understanding the controversy over the identity of EDRF. Nature. 1994;368:62- 65. 16. Palmer RJJ, Ferrige, A. G. & Moncada, S. Nitric oxide release accounts for the biological activity of endothelium-derived relaxing factor. Nature. 1987;327:524- 526. 17. Palmer RJJ, Ashton, D. S. & Moncada, S. Vascular endothelial cells synthesize nitric oxide from L-arginine. Nature. 1988;333:664-666. 18. Palmer RMJM, S. A novel citrulline-forming enzyme implicated in the formation of nitric oxide by vascular endothelial cells. Biochem Biophys Res Commun. 1989; 158:348-352. 100

19. Sessa WC. eNOS at a glance. J Cell Sci. May 15 2004;117(Pt 12):2427-2429. 20. Bryan NS BK, Murad F. Discovery of the nitric oxide signaling pathway and targets for drug development. Frontiers in Bioscience. 2009;14:1-18. 21. Hirafuji M, Tsunoda M, Machida T, et al. Reduced expressions of inducible nitric oxide synthase and cyclooxygenase-2 in vascular smooth muscle cells of stroke­ prone spontaneously hypertensive rats. Life Sci. Jan 11 2002;70(8):917-926. 22. Gladwin MT, Crawford JH, Patel RP. The biochemistry of nitric oxide, nitrite, and hemoglobin: role in blood flow regulation. Free Radie Biol Med. Mar 15 2004;36(6):707-717. 23. Feletou M, Vanhoutte PM. Endothelium-dependent hyperpolarizations: past beliefs and present facts. Ann Med. 2007;39(7):495-516. 24. Feletou M, Vanhoutte PM. EDHF: an update. Clin Sci (Lond). Aug 2009; 117(4): 139-155. 25. Andrews KL, Irvine JC, Tare M, et al. A role for nitroxyl (HNO) as an endothelium-derived relaxing and hyperpolarizing factor in resistance arteries. Br J Pharmacol. Jun 2009; 157(4):540-550. 26. Pagliaro P. Differential biological effects of products of nitric oxide (NO) synthase: it is not enough to say NO. Life Sci. Sep 12 2003;73(17):2137-2149. 27. Fukuto JM, Switzer CH, Miranda KM, Wink DA. Nitroxyl (HNO): chemistry, biochemistry, and pharmacology. Annu Rev Pharmacol Toxicol. 2005;45:335- 355. 28. Fukuto JM, Dutton AS, Houk KN. The chemistry and biology of nitroxyl (HNO): a chemically unique species with novel and important biological activity. Chembiochem. Apr 2005;6(4):612-619. 29. Paolocci N, Wink DA. The shy Angeli and his elusive creature: the HNO route to vasodilation. Am J Physiol Heart Circ Physiol. May 2009;296(5):H1217-1220. 30. Gratzel M, Taniguchi, S & Henglein, A. A pulse radiolytic sutdy of short-lived byproducts on nitric oxide reduction in aqueous solution. Ber Bunsenges Phys Chem. 1970;74:1003-1010. 31. Shafirovich V, Lymar SV. Nitroxyl and its anion in aqueous solutions: spin states, protic equilibria, and reactivities toward oxygen and nitric oxide. Proc Natl Acad Sci US A. May 28 2002;99(11):7340-7345. 32. Bartberger MD, Fukuto, J.M., & Houk, K.N. On the acidity and reactivity of HNO in aqueous solution and biological systems. Proc Natl Acad Sci U S A. 2001 ;98:2194-2198. 33. Bartberger MD, Liu W, Ford E, et al. The reduction potential of nitric oxide (NO) and its importance to NO biochemistry. Proc Natl Acad Sci US A. Aug 20 2002;99(17):10958-10963. 34. Gil'yano NY, Stepanov SI, Konevega LV, Noskin LA. Intracellular NO concentration and its changes in carcinoma cells and cultured human endotheliocytes under the influence of inhibitors and inductors of nitric oxide synthases. Bull Exp Biol Med. Feb 2007;143(2):248-250. 35. Hobbs AJ, Fukuto JM, Ignarro LJ. Formation of free nitric oxide from I-arginine by nitric oxide synthase: direct enhancement of generation by superoxide dismutase. Proc Natl Acad Sci US A. Nov 8 1994;91(23):10992-10996. 101

36. Schmidt HH, Hofmann H, Schindler U, Shutenko ZS, Cunningham DD, Feelisch M. No .NO from NO synthase. Proc Natl Acad Sci U S A. Dec 10 1996;93(25): 14492-14497. 37. Adak S, Wang Q, Stuehr DJ. Arginine conversion to nitroxide by tetrahydrobiopterin-free neuronal nitric-oxide synthase. Implications for mechanism. J Biol Chem. Oct 27 2000;275(43):33554-33561. 38. Miranda KM. The chemistry of nitroxyl (HNO) and implications in biology. Coordin. Chem. Rev. 2005;249:433-455. 39. Irvine JC, Ritchie RH, Favaloro JL, Andrews KL, Widdop RE, Kemp-Harper BK. Nitroxyl (HNO): the Cinderella of the nitric oxide story. Trends Pharmacol Sci. Dec 2008;29(12):601-60'8. 40. Paolocci N, Jackson MI, Lopez BE, et al. The pharmacology of nitroxyl (HNO) and its therapeutic potential: not just the Janus face of NO. Pharmacol Ther. Feb 2007; 113(2):442-458. 41. Amelie D.R. SJS. NO+, NO·, and NO- donation by S-nitrosothiols: implications for regulation of physiological functions by S-nitrosylation and acceleration of disulfide formation. Arch Biochem Biophys. 1995;318:279-285. 42. Tocchetti CG, Wang W, Froehlich JP, et al. Nitroxyl improves cellular heart function by directly enhancing cardiac sarcoplasmic reticulum Ca2+ cycling. Circ Res. Jan 5 2007;100(1):96-104. 43. Cheong E, Tumbev V, Abramson J, Salama G, Stoyanovsky DA. Nitroxyl triggers Ca2+ release from skeletal and cardiac sarcoplasmic reticulum by oxidizing ryanodine receptors. Cell Calcium. Jan 2005;37(1):87-96. 44. Lopez BE, Wink DA, Fukuto JM. The inhibition of glyceraldehyde-3-phosphate dehydrogenase by nitroxyl (HNO). Arch Biochem Biophys. Sep 15 2007;465(2):430-436. 45. Flores-Santana W, Switzer C, Ridnour LA, et al. Comparing the chemical biology of NO and HNO. Arch Pharm Res. Aug 2009;32(8):1139-1153. 46. Stasch JP, Schmidt PM, Nedvetsky PI, et al. Targeting the heme-oxidized nitric oxide receptor for selective vasodilatation of diseased blood vessels. J Clin Invest. Sep 2006; 116(9):2552-2561. , 47. Fukuto JM, Chiang K, Hszieh R, Wong P, Chaudhuri G. The pharmacological activity of nitroxyl: a potent vasodilator with activity similar to nitric oxide and/or endothelium-derived relaxing factor. J Pharmacol Exp Ther. Nov 1992;263(2):546-551. , 48. Liochev SI, Fridovich I. Nitroxyl (NO-): a substrate for superoxide dismutase. Arch Biochem Biophys. Jun 15 2002;402(2):166-171. 49. Fukuto JM, Bianco CL, Chavez TA. Nitroxyl (HNO) signaling. Free Radie Biol Med. Nov 1 2009;47(9):1318-1324. 50. Miranda KM, Paolocci N, Katori T, et al. A biochemical rationale for the discrete behavior of nitroxyl and nitric oxide in the cardiovascular system. Proc Natl A cad Sci USA. Aug 5 2003;100(16):9196-9201. 51. Liochev SI, Fridovich I. The mode of decomposition of Angeli's salt (Na2N2O3) and the effects thereon of oxygen, nitrite, superoxide dismutase, and glutathione. Free Radie Biol Med. Jun 1 2003 ;34(11 ): 1399-1404. 102

52. Miller TW, Cherney MM, Lee AJ, et al. The effects of nitroxyl (HNO) on soluble guanylate cyclase activity: interactions at ferrous heme and cysteine thiols. J Biol Chem. Aug 14 2009;284(33):21788-21796. 53. Paolocci N, Saavedra WF, Miranda KM, et al. Nitroxyl anion exerts redox­ sensitive positive cardiac inotropy in vivo by calcitonin gene-related peptide signaling. Proc Natl Acad Sci US A. Aug 28 2001;98(18):10463-10468. 54. 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. May 2009;296(5):H1274-1280~ 55. Irvine JC, Favaloro JL, Kemp-Harper BK. NO- activates soluble guanylate cyclase and Kv channels to vasodilate resistance arteries. Hypertension. Jun 2003;41(6): 1301-1307. 56. Favaloro JL, Kemp-Harper BK. The nitroxyl anion (HNO) is a potent dilator of rat coronary'vasculature. Cardiovasc Res. Feb 1 2007;73(3):587-596. 57. Yuill KH, Yarova P, Kemp XHBK, Garland CJ, Dora KA. A Novel Role for HNO in Local and Spreading Vasodilatation in Rat Mesenteric Resistance Arteries. Antioxid Redox Signal. Oct 7 2010. 58. Switzer CH, Flores-Santana W, Mancardi D, et al. The emergence of nitroxyl (HNO) as a pharmacological agent. Biochim Biophys Acta. Jul 2009; 1787(7):835- . 840. 59. Pagliaro P. Nitroxyl affords thiol-sensitive myocardial protective effects akin to early preconditioning. Free Radie Biol Med. 2003;34:33-43. 60. Paolocci N, Katori T, Champion HC, et al. Positive inotropic and lusitropic effects of HNO/NO- in failing hearts: independence from beta-adrenergic signaling. Proc Natl Acad Sci US A. Apr 29 2003;100(9):5537-5542. 61. Irvine JC, Favaloro JL, Widdop RE, Kemp-Harper BK. Nitroxyl anion dopor, Angeli's salt, does not develop tolerance in rat isolated aortae. Hypertension. Apr 2007;49( 4):885-892. 62. Bullen ML, Miller AA, Andrews KL, et al. Nitroxyl (HNO) as a Vasoprotective Signaling Molecule. Antioxid Redox Signal. Nov 2 2010. 63. Irvine J, Kemp-Harper BK, Widdop RE. Chronic administration of the HNO donor, Angeli's salt does not lead to tolerance, cross-tolerance or endothelial dysfunction: comparison with GTN and DEA/NO. Antioxid Redox Signal. Sep 19 2010. 64. Bermejo E, Saenz DA, Alberto F, Rosenstein RE, Bari SE, Lazzari MA. Effect of nitroxyl on human platelets function. Thromb Haemost. Sep 2005;94(3):578-584. 65. Cirino G, Fiorucci S, Sessa WC. Endothelial nitric oxide synthase: the Cinderella of inflammation? Trends Pharmacol Sci. Feb 2003;24(2):91-95. 66. Daiber A, Oelze M, Wenzel P, et al. Nitrate tolerance as a model of vascular dysfunction: roles for mitochondrial aldehyde dehydrogenase and mitochondrial oxidative stress. Pharmacol Rep. Jan-Feb 2009;61(1):33-48. 67. Vanhoutte PM. How we learned to say NO. Arteriosclerosis, Thrombosis and Vascular Biology. 2009;29. 68. Vanhoutte PM. How We Learned to Say NO. Arterioscler Thromb Vase Biol. Aug 2009;29(8): 1156-1160. 103

69. Savoia C, Schiffrin EL. Inflammation in hypertension. Curr Opin Nephrol Hypertens. Mar 2006;15(2):152-158. 70. Pufahl RA, Wishnok JS, Marletta MA. Hydrogen peroxide-supported oxidation of NG-hydroxy-L-arginine by nitric oxide synthase. Biochemistry. Feb 14 1995;34(6): 1930-1941. 71. Sharpe M. CC. Reactions of nitric oxide with mitochondrial cytochrome c: a novel mechanism for the formation of nitroxyl anion and peroxynitrite . . Biochem J. 1998;332:9-19. 72. Ellis A, Li CG, Rand MJ. Differential actions of L-cysteine on responses to nitric oxide, nitroxyl anions and EDRF in the rat aorta. Br J Pharmacol. Jan 2000; 129(2):315-322. 73. Pino RZ, Feelisch M. Bioassay discrimination between nitric oxide (NO.) and nitroxyl (NO-) using L-cysteine. Biochem Biophys Res Commun. May 30 1994;201(1):54-62. 74. Mistry DK, Garland CJ. Nitric oxide (NO)-induced activation of large conductance Ca2+-dependent K+ channels (BK(Ca)) in smooth muscle cells isolated from the rat mesenteric artery. Br J Pharmacol. Jul 1998;124(6):1131- 1140. 75. Mistry DK, Garland CJ. Characteristics of single, large-conductance calcium­ dependent potassium channels (BKCa) from smooth muscle cells isolated from the rabbit mesenteric artery. J Membr Biol. Jul 15 1998;164(2):125-138. 76. Giachini FR, Carneiro FS, Lima VV, et al. Upregulation of intermediate calcium­ activated potassium channels counterbalance the impaired endothelium-dependent vasodilation in stroke-prone spontaneously hypertensive rats. Transl Res. Oct 2009; 154(4): 183-193. 77. Versari D, Daghini E, Virdis A, Ghiadoni L, Taddei S. Endothelium-dependent contractions and endothelial dysfunction in human hypertension. Br J Pharmacol. Jun 2009;157(4):527-536. 78. Briones AM, Touyz RM. Oxidative stress and hypertension: current concepts. Curr Hypertens Rep. Apr 2010;12(2):135-142. 79. Raines EW, Garton KJ, Ferri N. Beyond the endothelium: NF-kappaB regulation of smooth muscle function. Circ Res. Apr 2 2004;94(6):706-708. 80. Didion SP, Kinzenbaw DA, Schrader LI, Chu Y, Faraci FM. Endogenous interleukin- IO inhibits angiotensin II-induced vascular dysfunction. Hypertension. Sep 2009;54(3):619-624. 81. Savoia C, Schiffrin EL. Vascular inflammation in hypertension and diabetes: molecular mechanisms and therapeutic interventions. Clin Sci (Land). Jun 2007;112(7):375-384. 82. Ellis A, Lu H, Li CG, Rand MJ. Effects of agents that inactivate free radical NO (NO*) on nitroxyl anion-mediated relaxations, and on the detection of NO* released from the nitroxyl anion donor Angeli's salt. Br J Pharmacol. Oct 2001; 134(3):521-528. 83. Wanstall JC, Jeffery TK, Gambino A, Lovren F, Triggle CR. Vascular smooth muscle relaxation mediated by nitric oxide donors: a comparison with acetylcholine, nitric oxide and nitroxyl ion. Br J Pharmacol. Oct 2001; 134(3):463-4 72. 104

84. Miranda KM, Nims RW, Thomas DD, et al. Comparison of the reactivity of nitric oxide and nitroxyl with heme proteins. A chemical discussion of the differential biological effects of these redox related products of NOS. J Jnorg Biochem. Jan 1 2003;93(1-2):52-60. 85. Fukuto JM, Bartberger MD, Dutton AS, Paolocci N, Wink DA, Houk KN. The physiological chemistry and biological activity of nitroxyl (HNO): the neglected, misunderstood, and enigmatic nitrogen . oxide. Chem Res Toxicol. May 2005; 18(5):790-801. 86. Saleem M, Ohshima H. Xanthine oxidase converts nitric oxide to nitroxyl that inactivates the enzyme. Biochem Biophys Res Commun. Mar 5 2004;315(2):455- 462. 87. Donzelli S, Espey MG, Thomas DD, et al. Discriminating formation of HNO from other reactive nitrogen oxide species. Free Radie Biol Med. Mar 15 2006;40(6): 1056-1066. 88. Donzelli S, Switzer CH, Thomas DD, et al. The activation of metabolites of nitric oxide synthase by metals is both redox and oxygen dependent: a new feature of nitrogen oxide signaling. Antioxid Redox Signal. Jul-Aug 2006;8(7-8):1363-1371. 89. Amatore C, Arbault S, Ducrocq C, Hu S, Tapsoba I. Angeli's salt (Na2N2O3) is a precursor of HNO and NO: a voltammetric study of the reactive intermediates released by Angeli's salt decomposition. ChemMedChem. Jun 2007;2(6):898-903. 90. Miranda KM, Dutton AS, Ridnour LA, et al. Mechanism of aerobic decomposition of Angeli's salt (sodium trioxodinitrate) at physiological pH. J Am Chem Soc. Jan 19 2005;127(2):722-731. 91. Li CG, Karagiannis J, Rand MJ. Comparison of the redox forms of nitrogen monoxide with the nitrergic transmitter in the rat anococcygeus muscle. Br J Pharmacol. Jun 1999; 127( 4):826-834. 92. Guilmard C, Auguet M, Chabrier PE. Comparison between endothelial and neuronal nitr_ic oxide pathways in rat aorta and gastric fundus. Nitric Oxide. 1998;2(3): 147-154. 93. Yang HY, Kao PF, Chen TH, Tomlinson B, Ko WC, Chan P. Effects of the angiotensin II type 1 receptor antagonist valsartan on the expression of superoxide dismutase in hypertensive patients. J Clin Pharmacol. Mar 2007;47(3):397-403. 94. Misra HP. Generation of superoxide free radical during the autooxidation of thiols. J Biol Chem. 1974;219:2151-2155. 95. Saez G, Thomalley, P.J., Hill, H.A.O., Hems, R. and Bannister, J.V. The production of free radicals during the autoxidation of cysteine and their effect on isolated rat hepatocytes. Biochem Biophys Acta. 1982;719:24-31. 96. Jia L, Furchgott RF. Inhibition by sulfhydryl compounds of vascular relaxation induced by nitric oxide and endothelium-derived relaxing factor. J Pharmacol Exp Ther. Oct 1993;267(1):371-378. 97. Murrell W. Nitroglycerin as a remedy for angina pectoris. Lancet 1879;1 :80-81. 98. Frame MD, Fox RJ, Kim D, Mohan A, Berk BC, Yan C. Diminished arteriolar responses in nitrate tolerance involve ROS and angiotensin II. Am J Physiol Heart Circ Physiol. Jun 2002;282(6):H2377-2385. 99. Fukatsu A, Hayashi T, Miyazaki-Akita A, et al. Possible usefulness of apocynin, an NADPH oxidase inhibitor, for nitrate tolerance: prevention of NO donor- 105

induced endothelial cell abnormalities. Am J Physiol Heart Circ Physiol. Jul 2007 ;293 ( 1): H790-797. 100. Szocs K, Lassegue B, Wenzel P, et al. Increased superoxide production in nitrate tolerance is associated with NAD(P)H oxidase and aldehyde dehydrogenase 2 downregulation. J Mo! Cell Cardiol. Jun 2007;42(6):1111-1118. 101. Kosmicki MA. Long-term use of short- and long-acting nitrates in stable angina pectoris. Curr Clin Pharmacol. May 2009;4(2): 132-141. 102. Schiffrin EL. The vascular phenotypes in hypertension: Relation with the natural history of hypertension. J Am Soc Hypertens. Jan-Feb 2007;1(1):56-67. 103. Lima VV, Giachini FR, Carneiro FS, et al. O-GlcNAcylation contributes to the vascular effects of ET-1 via activation of the RhoA/Rho-kinase pathway. Cardiovasc Res. Nov 21 20i0. 104. Poliseno L, Cecchettini A, Mariani L, et al. Resting smooth muscle cells as a model for studying vascular cell activation. Tissue Cell. Apr 2006;38(2):111-120. 105. Schulz E, Tsilimingas N, Rinze R, et al. Functional and biochemical analysis of endothelial (dys)function and NO/cGMP signaling in human blood vessels with and without nitroglycerin pretreatment. Circulation. Mar 12 2002; 105(10): 1170- 1175. 106. Martin W. Nitroxyl anion--the universal signalling partner of endogenously produced nitric oxide? Br J Pharmacol. Jun 2009;157(4):537-539. 107. Munzel T SH, Freeman BA, Tarpey MM, Harrison DG. Evidence for enhanced vascular superoxide anion production in nitrate tolerance. A novel mechanism underlying tolerance and cross-tolerance. J Clin Invest. 1995 ;95: 187-194. 108. Dikalov S, Fink B, Skatchkov M, Sommer 0, Bassenge E. Formation of Reactive Oxygen Species in Various Vascular Cells During Glyceryltrinitrate Metabolism. J Cardiovasc Pharmacol Ther. Jan 1998;3(1):51-62. 109. Sage PR, de la Lande IS, Stafford I, et al. Nitroglycerin tolerance in human vessels: evidence for impaired nitroglycerin bioconversion. Circulation. Dec 5 2000; 102(23):2810-2815. 110. Munzel T, Kurz S, Rajagopalan S, et al. Hydralazine prevents nitroglycerin tolerance by inhibiting activation of a membrane-bound NADH oxidase. A new action for an old drug. J Clin Invest. Sep 15 1996;98(6):1465-1470. 111. Mc Veigh GE, Hamilton P, Wilson M, et al. Platelet nitric oxide and superoxide release during the development of nitrate· tolerance: effect of supplemental ascorbate. Circulation. Jul 9 2002;106(2):208-213. 112. Otto A, Fontaine J, Tschirhart E, Fontaine D, Berkenboom G. Rosuvastatin treatment protects against nitrate-induced oxidative stress in eNOS lmockout mice: implication of the NAD(P)H oxidase pathway. Br J Pharmacol. Jun 2006; 148(4):544-552. 113. Zhu S, White RE, Barman SA. Effect of PKC isozyme inhibition on forskolin­ induced activation of BKCa cp.annels in rat pulmonary arterial smooth muscle. Lung. Mar-Apr 2006;184(2):89-97. 114. White RE, Kryman JP, El-Mowafy AM, Han G, Carrier GO. cAMP-dependent vasodilators cross-activate the cGMP-dependent protein kinase to stimulate BK(Ca) channel activity in coronary artery smooth muscle cells. Circ Res. Apr 28 2000;86(8):897-905. 106

115. White RE, Darkow DJ, Lang JL. Estrogen relaxes coronary arteries by opening BKCa channels through a cGMP-dependent mechanism. Circ Res. Nov 1995;77(5):936-942. 116. Carrier GO, Fuchs LC, Winecoff AP, Giulumian AD, White RE. Nitrovasodilators relax mesenteric microvessels by cGMP-induced stimulation of Ca-activated K channels. Am J Physiol. Jul 1997;273(1 Pt 2):H76-84. 117. Fukuto JM, Gulati P, Nagasawa HT. Involvement of nitroxyl (HNO) in the cyanamide-induced vasorelaxation of rabbit aorta. Biochem Pharmacol. Mar 2 1994;4 7(5):922-924. 118. Munzel T, Kurz S, Heitzer T, Harrison DO. New insights into mechanisms underlying nitrate tolerance. Am J Cardiol. May 30 1996;77(13):24C-30C. 119. Wenzel P, Mollnau H, Oelze M, et al. First evidence for a crosstalk between mitochondrial and NADPH oxidase-derived reactive oxygen species m nitroglycerin-triggered vascular dysfunction. Antioxid Redox Signal. Aug 2008; 10(8): 1435-1447. 107

APPENDIX

Vascular smooth muscle cell signaling mechanisms for contraction to angiotensin II and endothelin-1. Wynne, B. M., Chiao, Chin-Wei, Webb, R. C. Journal of American

Society ofHypertension. 2008

Mesenteric arteries from spontaneously hypertensive stroke prone rats exhibit an increase in NO-dependent vasorelaxation via iNOS. Wynne, B. M., Giachini, F. R., Labazi, H.,

Lima, V. V., Carneiro, F., Dorrance, A. M., Webb, R. C., Tastes, R. C. Journal of the American Society of Hypertension ■( ■) (2008) 1-13 58 2 Review Article 59 3 60 4 Vascular smooth muscle cell signaling mechanisms for contraction 61 5 ·62 6 to angiote~sin II and endothelin-1 63 7 64 8 Brandi M. Wynne, MS*, Chin-·Wei Chiao, PhD, and R.' Clinton_Webb, PhD 65 9Q1 Department of Physiology, Medical College of Georgia, Augusta, Georgia, USA 66 10 Manuscript rece~ved June 28, 2008 and accepted September 75, 2008 67 11 68 12 69 13 Abstract 70 14 71 15 Vasoactive peptides, such as endothelin-1 and angiotensin II, are recognized PY~:§pecific receptor proteins ·located in the cell 72 16 membrane of target cells. After receptor recognition, the spec:;ificity of the cell'ular r~sponse is achieved by G-protein coupling 73 17 of ligand binding to the regulation of intracellular effectors. These intracellti'tar•. ~f:fbctors will be_ the subject of this brief re­ 74 18 view on contractile activity initiated by endothelin-1 and angiotensin II.,,i2lit,y~tion of receptors by endothelin-1 and angio­ 75 19 tensin Il·in smooth muscle cells results in phospholipase C activation •l~~g.ing.to the generation of the second messengers 76 20 2 inositol trisphosphate (IP3) and diacylglycerol (DAG). IP3 stimulatesj:n:ty.c·~1:1u1ar Ca + release from the sarcoplasmic retic­ 77 ·21 ulum and DAG causes protein kinase C activation. Additionally, diffi".'. · Ca2+ entry channels, such as voltage-operated, re­ 78 22 ceptor-operated, and store-operated Ca2+ channels, as well as Ca2+-pe -:able nonselective cation channels, are involved in 79 '23 the elevation of intracellular Ca2+ C(?ncentration. The elevation 1~,;;i~~~cellular Ca2+ is transient and initiates contractile ac­ 80•, 24 tivity by a Ca2+-calmodulin interaction, stimulating myosin ligp~ chani (MLC) phosphorylation. When the Ca2+ concentra- 81 25 tion begins to decline, Ca2+ sensitization of the contractile t€ln:'S;J/ signaled _by the RhoA/Rho-kinase pathway to inhibit 82 26 the dephosphorylation of MLC phosphatase (MLCP), ther ' ·iµ.aining force generation. Removal of Ca2+ from the cy- 83 27 tosol and stimulation of MLCP initiates the process of sm~ ;~tle relaxation. In pathologic conditions such as hyperten- 84 28' sion, alterations in these cellular signaling compo~ts., cart lead to an overstimulated · state causing maintained 85 29 vasoconstriction and blood pressure elevation. J A1I1:·;:::;oc··H,ypertens 2008; ■(■ ):l-13. © 2008 American Society of 86 30 Hypertension. AU rights reserved. · '\ 87 31 Keywords: Calcium channel; Rho-kinase; vasoactive peptides; P,pciip}!Q.i~9iitides. 88 32 ,;=; -:..:.··· 89 33 90 34 91 35 Introduction Arteries are composed of three layers: 1) the tunica: ad­ 92 36 ventitia, 2) tunica media, and 3) tunica intima. Although 93 In the vasculature, the small arteries_,,.,a~d"'';arterioles the outer and inner layers are composed mainly of connec- 37 regulate the majority of blood flow resist~ceyfiifthe circu­ 94 38 1 ~ive tissue and endothelial cells, respectively, the tunica me­ 95 lation. Circulating neurotransmitters, h,9xqiHh~.,ey. endothe­ dia is composed of smooth muscle cells.3 All smooth 39 lium-de:ived facto:s, and even sheaf· strt1 -itself plays 96 40 muscle cells, regardless of the stimulus, produce force or. 97. a role m modulatmg smooth muscie~,-~on~l and, conse­ contraction . through cross-bridge cycling between actin 41 1 3 98- quently, lumen diameter. - The fi.n;11'fprgfltl~t is the control 42 .iJ' .... and myos'in filaments. Although vascular smooth muscle 99 of blood pressure (BP) and organ 1?.~ood "!ilow. cells are capable of dynamic changes in gene expression 43 •-:~·,:,-: ,,.,/. 100 to either contractile or differentiation proteins, the contrac­ 44 4 101 45 tile phenotype in vascular. smooth muscle predominates. 102 46 The contractile response of vascular smooth muscle is the 103 product of myosin light chain kinase (MLCK) and myosin 47 This study was supporte~. :by !i~tional Institutes of Health 104 1 light chain phosphatase (MLCP) activation. In smooth mus­ 48 (Grant Nos. HL71138 and HL141.i69-). 105 cle, this process is initiated by a c_alcium (Ca2+)-mediated 49 Conflict of interest: none. 2 106 change in the thick (myosin) filaments. · 50 *Corresponding author: Brandi M. Wynne, MS, Department of 107 51 Physiology, Medical College of Georgia, Augusta, Georgia 30912. In this review, we will discuss the moiecular mechanisms 108 52 Tel: 706-721-3547; fax: 706-721-7299. by which two common vasoactive peptides, angiotensin II 109 E-mail: [email protected] 53 and endothelin-1, produce smooth muscle contraction. 110 54 1933-1711/08/$ - see front matter © 2008 American Society of Hypertension. All rights reserved. 111 55 doi:10.1016/j.jash.2008.09.002 112 56 113 57 114

Dli'V ,;: n nTn • TA~Hl '.\Ii nrnnf ■ ?.Q Octnher 2008 ■ 3:57 am ■ ce 2 B.M. Wynne et al. I Journal of the American Society of Hypertension ■ ( ■ ) (2008) 1-13

1-15 Because intracellular Ca2+ is fundamental to the contractile phosphatases. This receptor isoform has also been docu- 172 116 process, we will also provide a brief review of the cellular mented to trigger the kinin-NO-cGMP pathway. 8 However, 173 117 mechanisms that regulate this cation. it is through the AT 1 receptor with which the majority of 174 118 physiologic and pathologic actions of angiotensin II is me- 175 119 diated, and will be the focus of the signaling in this review. 176 Angiotensin II Signaling 120 The angiotensin II signaling cascade via the AT 1 receptor 177 121 Angiotensin II, which is the predominate bioactive pep- is a consequence of the particular G-protein or signaling 178 122 5 6 tide in the renin-angiotensin system (RAS), promotes the cascade that is activated Gq11 1, Gi, G12, and G13 . • The 179 123 maintenance of systemic pressure through various mecha- AT1 receptor signals through three main pathways: classical 180 124 nisms in the cardiovascular and renal systems.5 Although phospholipase C (Pi,C). signaling leading to inositol tri- 181 125 angiotensin II is crucial to salt and water homeostasis, sphosphate (IP3) a~cl' ~jacylglycerol (DAG) cleavage, phos- 182 126 this peptide is also implicated in several cardiovascular pholipase D, also'J·,~jr.1inating in DAG generation; and 183 127 conditions, such as hypertension, atherosclerosis, and heart phosphatidylch~iii'I&1~tJ?,hdsphatidic acid and phospholipase 184 128 failure, and more specifically in vascular smooth muscle A2, which is r~;~onsi)l~ for arachidonic acid and prosta- 185 129 contraction; vascular remodeling including the induction glandin proqµ,,<}tjphr;,•,'-fyrosine phosphorylation, ie, mitogen 186 130 of hypertrophy and hyperplasia. Canonical angiotensin II activated pr,oteiii_'?!pnase (MAPK), which is distinctive· for 187 131 signaling occurs through a membrane bound heterotrimeric its role in g~g~ttlfactor and cytokine activation, has been 188 132 G-protein coupled receptor (GPCR). To date, there are two associa~¢'B~tth"'increased levels of AT1 receptor activation. 189 133 GPCRs known to mediate angiotensin II function: the AT 1 Nonreei,Vi.te:)rtyfosine kinase activity has ·also been associ- 190 134 6 and AT2 receptors.5• . . atedlw,i,~rangiotensin II, including but not limited to: the 191 135 The AT 1 receptor is expressed in a variety of tissues, in- Src(J¥hlly of kinases, praline-rich tyrosine kinases 192 136 eluding the kidney, heart, adrenal gland, brain, lung, and (Pyksj;::,e~tracellular signal-regulated kinases (ERKs), pax- 193 137 adipose; however, the focus will be on AT receptor activa~ ,illi:111 focal adhesion kinase, phosphoinositide 3 kinase and 194 138 1 tion in arterial smooth muscle. The AT2 receptor is highly .. · an ihflammatory cytokine pathway, janus kinases/signal 195 139 expressed in fetal tissues, then declining quickly after birth. · 'ttan~oucers and. activators of transcription.7 Angiotensin 196 140 In adults, the AT2 receptor has a varied tissue distribution as/: , · II js a pleiotropic signaling molecule, whose varied path- 1Q7 141 well, and is detectable in the kidney and adrenals, pancrec1:~;' ·· : '·f~\~o/-ays is complex and specific, but yet also converges to 198 142 ovary, brain, heart, and vasculature. In vessels, the AT2.J_M~::,:;, Bring forth multiple responses. This review will focus on 199 143 ceptor is not localized on the smooth muscle, but is maj,ply the PLC pathway, culminating in vascular smooth muscle 200 144 expressed in the adventitia. After injury, levels ·of t1fs re.:.:,'.' contraction. 201 145 ceptor have been observed to increase, which may.,,d'p>J~i.9,:,/ Recent literature suggests that. the AT 1 receptor couples 202 146 current dogma suggesting that the AT2 receptor isrln ved to Gq to mediate vasoconstriction, which is the pathway 203 147 in cellular growth. In contrast to the vasoconstric "' ei{s leading to smooth muscle contraction via PLC. 9 In addition, 204 148 of the AT 1, the AT2 receptor has'bee~ showntRY~ ate angiotensin II was found to elicit a contractile response via 205 149 7 vasodilation. Less is known about AT2 re1rf~#"·irlbcha- a~tiva~on of G12'G13 G-proteins and the RGS-RhoGEF 206 150Q2 · nisms than the AT 1 receptor; however, it !,S:,k,116-Wa that it s1gnalmg pathway. Recently, insights into this mechanism Q3 207 151 couples to Gi and activates tyrosine anc;!{secl¾'"t~lthreonine were discovered and will be discussed in a later 208 152 ~~i W9 •<:;·~~ 153 210 154 phosphorylated 211 155 myosln light chain 212 156 213 157 214 158 myosin light chain 215 159. mysoin light chain kinase 216 160 phosphatase . 217 161 \ 218 162 Ca2+ -calmodulin 219 163 myosin light chain 220 164 (Inactive) . 221 165 222 Figure 1. Overview of major contributors to vascular smooth muscle contraction. In smooth muscle, contraction is initiated by a Cai+_ 166 mediated change in the thick filaments, or myosin. With myosin light chain phosphorylation, the actin and myosin filaments are capable 223 167 of interacting. ATP hydrolysis is the source for force generation in smooth muscle; with contraction, inorganic phosphate leaves the 224 168 myosin head. Relaxation occurs via the action of myosin light chain phosphatase, which dephosphorylates myosin light chain, inactivat­ 225 169 ing it. ATP, adenosine-5'-triphosphate; Ca2+, calcium. 226 170 227 171 228

1n.·v '- o nTn • JASH136 nroof ■ 29 October 2008 ■ 3:57 am ■ ce B.M. Wynne et al. I Journal of the American Society of Hypertension ■ ( ■ ) (2008) 1-13 3

229 286 230 287 231 288 232 289 233 290 234 291 235 292 236 293 237 294 2 238 MLKC Ca + -calmodulin . 295 239 - 296 240 297 241 ATP p 298 242 I 299 243 ~ 300 244 .. ~/(c~mraction) 301 245 (remxaflon~ 302 p-MLC 246 'MLC 303 247 (inactive) 304 248 l 305 249 MLCP 306 250 Rho-kinase __.. 307 251 !f 308· 252 p-MLCPCE§)1-- CPl-17 - PKC 309 253 (Inactive) f 310 254 311 ..tr j 255 312 Figure 2. Molecular mechanisms of smooth muscle contractiom::'"'!: . ~ smooth muscle contraction is the summation of MLCK and 313 256 MLCP activity. With receptor bi_nding, an increase in intrace~tdi- , a + occurs, both via channels located in the membrane and intracel­ 257 lular stores in the sarcoplasmic reticulum. The Ca2+ inter~i~ w1tH modulin, forming a Ca2+ -calmodulin complex, which activates 314 258 MLCK. MLCK can then MLC-P; allowing for the close iifteractt9n of the actin and myosin filaments for force generation; Relaxation 315 259 occurs with MLCP dephosphorylating MLC. In some inst~ces,. .fbrce generation can be tonic; this is mediated in a Ca2+ -independent 316 260 manner. Rho-kinase becomes activated via the small,,;whctt.;itetfRhoA protein, which subsequently phosphorylates MLCP, rendering 317 261 the enzyme inactive and incapable of de-phosphoryl!!i.'Ung'M4C. In addition, PKC and Rho kinase work in concert to activate CPI-17, 318 2 262 which inhibits MLCP. Thus, the vascular smooth m~~.sJe'tkQP,'.ot relax and a tonix contraction occurs. Ca +, calcium; MLCK, myosin light 319 chain kinase; MLCP, myosin light chain phosphatisdMiLC:-P, phosphorylate myosin light chain. 263 ,:_;·i, ,.r 320 264 321 265 10 11 2 14 section. • After Gq!G 11 GPCR activatiih, :tseco~d mes­ levels from the extracellular space. • The EF hand · 322 266 sengers are generated, beginning with "''F~C activation. Ca2+-bindiI).g protein, calmodulin, is the primary target 323 267 PLC is a membrane-bound enzyme res~·orlsiQJ~'l·or cleaving for elevated intracellular Ca2+. The Ca2+ -calmodulin com- 324 268 membrane lipid phosphoinositide 4, 5-:'~}>isphojphate, result­ plex is then able to activate MLCK.2 It is also or impor- . 325 269 2 2 ing in the generation of IP3 and DAg;, :'tZ9{'and DAG reg­ tance to note that the Ca +-tension relationship changes 326 270 ulate two distinct but parallel pathways l~~ding to increased for stimulation type and during the time of the contraction. 327 271 intracellular Ca2+ and ,9plmipat1ng in MLC In general, studies show that the receptor-mediated stimula- 328 272 phosphorylation.2•13 ·14 ·•;.• .... tions produce a greater tension for a given Ca2+ concentra- 329 273 IP3 regulates calcium relea'.se.;.j~to· the cytosol from the tion. Moreover, during the sustained phase of contraction, 330 274 sarcoplasmic reticulum (SR);'·R),edi:ated via the ryanodine lower levels of Ca2 + are needed, which is referred to as 331 275 receptor (RyR). The RyR,,"w.hich')as been shown to be lo­ Ca2+ sensitization. These are two differential points of reg- 332 276 calized centrally as well ai;;·"Pehpherally on the SR, is ulation for vascular smooth muscle contraction. 17 333 277 15 16 thought to facilitate optimal binding of IP3. • Activation DAG mediates vascular smooth muscle contraction via 334 278 of the RyR opens Ca2+ channels, causing· a rapid increase activation of protein kinase. C (PKC). PKC is a serine/thre- 335 279 in Ca2+ concentration. After cytosolic Ca2+ concentrations onine kinase known to exist in several'isoforms, all which 336 280 peak and then decline, there is a sustained elevation of contribute to various physiological activities and are acti- 337 281 Ca2+ above that of basal levels.2 This sustained increase vated differentially. Of the three known vascular smooth 338 282 in Ca2+ is due to receptor-operated Ca2+ channels on muscle isoforms, PKCa and PKC/3 are dependent on 339 283 plasma membrane facilitating increased cytosolic Ca 2+ DAG activation and intracellular Ca2+ concentrations, 340 284 341 285 342

nvu " n nTn • TA '"l-l' 1 '.l/; nrnnf • ?Q n..,tnh,-r ?00R ■ 3:57 am ■ ce 4 B.M. Wynne et al. I Journal of the American Society of Hypertension ■ ( ■ ) (2008) 1-13

343 400 0 1 344 ;:.r.t;.. i'.-_.,·~.·.·.,!2.:.: ..'~·rt· ...·,.;,-G_:ni,~-e:.•.·- •... .l:.. :· •.. ~.·.:.·.•_;,:W:·;•·· ...•.a:_~ ...·•.-n·•_-.:_~n.'..:_· .. e· .. 1'.· : ·sett' · 401 345 WH~ _, .WR :q.~·~~i~t 402 346 403 347 404 348 405 349 ~sma membrane _'.~,~~~;i;' =i/Jfi~J~~==~J~Ca•.~ 406 350 ~i~i• 407 351 PIP2 DAG \ ,: • / ~--- t\ ~ 408 352 . ~ . .,, , ,/ / Cl· \ l 409 353 PKC•' \ ,/. @:5 0/ 410 IP ', .• , @,, - 354 411 3 ~<0*0 355 ~0 ~u~ 412 356 413 357 414 358 415 359- Sarcoplasm;c reticulum[~ I~ =1 /"(--, I 416 membrane ~ U ~ / Ca-• depletion~ 360 0~~ 417 361 0~ (@) 418 . Qa2• store ,..

362 ,A.~r"•'~ ... 419 363 420 Figure 3, The concept of ROC Ca2+, VOC Caz+, and SOC Ca2+. ROC chal~J,f~e activated via receptor stimulation of GPCRs directly 364 421 or through the production of s~cond messengers such as IP3 or DAG. IP3 can theWbind to and activate the IP3 receptor on the sarcoplasmic 365 reticulum membrane, causing discharge and release of the stored Ca2+. ::Uhe-~lease of Ca2+ from the sarcoplasmi~ reticulum induces Cl_ 422 . 366 423 efflux or the influx of Na+ and Ca+ from the ROC channels, which emf then !f~tivate VOC channels. The second messenger, IP3 activates 2 2 367 the IP3 receptor on the sarcoplasmic reticulum membrane, causing dls~ffar,g~ ahd release of the stored c~ +. After Ca + depletion from the 424 368 sarcoplasmic reticulum,·the SOC channels are activated. A, agol}ilit;.~a2+/calcium; DAG, diacylglycerol; GPCRs, G protein-coupled 425 369 receptors; IP3, inositol 1,4,5-trisphosphate; IP3R, IP3 receptor; J;>ll?i;,':ph<:i~phatidylinositol 4,5-bisphosphate; PLC, phospholipase C; R, 426 370 receptor; ROC, receptor-operated channels; TK, tyrosine kinas~ ..·soc:·itore-operated channels; VOC, voltage-operated channels. . 427 ., . '~. '• 371 428 372 whereas PKCE is DAG-dependent only. PKC modµIites·•· : subsequent actin-myosin cross-bridging, but the process· 429 373 vascular smooth muscle contraction by directly phosphpry~ .. of force maintenance is mediated in a Ca2+ -independent 430 2 18 22 374. lating MLCK, which subsequentiy phosphorylat¢i Mi:"c'.· ' manner. • • MLCP is the contender, regulating the inacti­ 431 375 Furthermore, PKC activation leads to the acti~aJ~bn.~of vation- of MLC by removing the high-energy phosphate , 432 376 several other target proteins promoting- smo,@t9. .·rriuscle group, promoting smooth muscle relaxation. MLCP is com­ 433 377 contraction. PKC phosphorylates and activ;i~s/ER>Kl/2, posed of two subunits; one _is the catalytic 38 kDa subunit 434 378 Rho-kinase (p 160ROCK), and calmod,\;llfrtltl~pendent of type 1 protein phosphatase (PPlc, o isoform) and two 435 12 17 379 protein kinase II. In addition, intracellul~;,i C.#1't·.elevation other noncatalytic subunits. • This mechanism generates 436 380 was suggested to regulate angiotensin-II-in:Utt~,~-d epidermal force maintenance by means of Ca2+ -sensitization of the 437 . 381 growth factor receptor and ERK activ,at'i'dh:i: 1· 'Various ion contractile proteins. This is signaled by the RhoA/Rho ki- - 438 . . ~~ ·1. 382 channels and ion transporters have b't;,~n dep:ionstrated as nase pathway, which inhibits the dephosphory-lation of 439 383 downstream targets of PKC. This.,-ph~ifo5roJnon has been MLC by MLCP. RhoA is a small, monomeric G protein, 440 384 illustrated with the use of the sp!_cific lKC agonists (ie, which is regulated by the binding of GTP. This process is 441 385 phorbol esters) which induce s~q~h:,-,W,Ji'Scle contraction.2 facilitated by nucleotide exchange factors, RhoGEFs, 442 ·443 386 All the earlier events described'1'0ulminating in MLCK which enable the exchange of GDP for GTP, activating Q4 387 activation, resulting in contr .,,,,,.;,,,.:-.;l(;tftthe phosphorylation RhoA. Upon RhoA activation, the RhoA-GTP complex is 444 388 of Ser19 of the 20 k.Da re i,, ':·protein of MLC. The able to activate Rho kinase, a serine/threonine kinase. 445 389 phosphorylation.of MLC alll?,;VS f~r the interaction between. Rho kinase can then phosphorylate the myosin-binding 446 390 actin and myosin filaments. ·tntrinsic adenosine-5 '-triphos­ subunit of MLCP, inhibiting it, and thus preventing the de­ 447 391 phatase (ATPase) actiyity of myosin is crucial for this pro­ phosphorylation of MLC, causing maintenance of contrac­ 448 2 23 24 392 cess, because the hydrolysis and subsequent release of ATP tion. • · In addition to the re-gulatory effects of Rho 449 393 results in the cycling of the myosin cross-bridges with kinase, CPI-17 is another protein· that regulates MYPT Qs 450 8 21 394 actin, causing force generation.3• 1 - - phosphorylation status by inhibiting PPlc, thus inactivating 451 395 Smooth muscle cells also contain Ca2+ -independent MLCP.. CPI-17 phosphorylation occurs in a Ca2+ -depen­ 452 396 mechanisms to regulate contractility. The process of force dent and Ca2+ -independent mechanism. Ca2+ -dependent 453 397 generation is mediated by . MLCK a~tivation, · and PKC was suggested to regulate. the rapid phosphorylation, 454 398 455 399 456 B.M. Wynne et al. I Journal of the American Society of Hypertension ■ ( ■ ) (2008) 1-13 5

457 514 store-operated Ca2+ 458 515 channel 459 516 460 Na+, ca2+ 517 461 518 462 519 463 · 520 464 521 465 522 466 523 467 524 468 525 ROCK 469 526 470 527 471 / MLC /JI~=====,::=;; 528 472 MLCP - if ~ MLCK I 529 473 MLC·® 530 474 SR 531 475 m_~mbrane 532 476 Figure 4. New pathways in vascular smo~th muscle signaling. Orail and S~·-,$6:t-k together to function as Ca2+ sensors within the 533 477 cell. Orail makes up the pore of the CRAC channel, allowing for Ca2+ ent(:~1i_~hereas STIMl senses Ca2+ levels within the SR. Their 534 478 interaction facilitates this process. The !Kea channels are intermediate concftfct@ce Ca2+-activated K+ channels which are of impor­ 535 479 tance in small resistance vessels. 0 12113 sigtiai.ing'thrbughLARG was rec~IJ.tly fou~d to be a mechanism for RhoA/Rho kinase activation, 536 480 leading to MLCP inactivation. Ca2+, calcium; CRAC, Ca2+ -release)i~-ti~ited Ca2+; !Kea?; MLCP, myosin. light chain phosphatase; 537' 481 STIM, stromal-interacting molecule. 538 482 539

483 whereas phosphorylation by PKC and Rho kinase occur_~•,. ·•· .1A~:ndothelin-l is the main isoform secreted by the end_o­ 540 484 independently of Ca2+ in the later phase. This. has b~e°';,. thelium, and has been shown to act in a paracrine or auto­ 541 485 demonstrated with the use of Rho kinase and PKC inhibi- •;' ".·.,, crine manner on its receptors in vascular smooth muscle. 542 486 tors; however, the effects were dependent on the type §{~g-·. . There are three known endothelin-1 receptors that have 543 487 onist and tissue used. Its activity and expression is a 'f!!,qto~.} an assorted tissue distribution and functional role: ETA 544 488 in the contractile state of vascular smooth muscl~>S'ancf"iifs"' · and ETB are present in mammals and a putative ETc in 545 489 also been shown to be of importance in Ca2+/: ·zi- non-mammals. The typical receptor found on vascular 546 12 17 490 tion. • Much needs to be explored regardin ~£~g- smooth muscle cells is the ETA receptor, which mediates 547 4 25 28 491 ulation, but regardless, Rho kinase and the vasoconstrictor ·effects of endothelin-1. • • Endothe- 548 492 acknowledged to be of importance in.vascu!~,. lin-1 receptor activation can lead to diverse responses in 549 493 cle contraction and force maintenance. ,i_:' }·,b,,,, the cell through interaction with pathways that are both per- 550 494 .,,,,,,._..,,- tussis toxin-sensitive and toxin-insensitive, leading to the 551 495 552 Endothelin-1 Signaling ::~"•;:;,i~;:'.•;.;);, conclusion that endothelin-1 acts through several GPCRs. 496 ;~~. Generally ETA receptors have been associated with vaso- 553 554 497 Endoth.elin-1, which is a 21-c!,µu.~k,,a:efd peptide, is constriction and cell growth, whereas ET8 receptors are in- 498 known as one of the most potent epdoge\ous vasoconstric- valved in the clearance of endothelin-1, inhibition of 555 25 27 499 tors as described by Yanagisawa.,~Ca~,i.P;,,1'988. - Since its endothelial cell apoptosis, the release of NO and prosta- 556 500 discovery, the endothelin-1 system11tas:become increasingly glandins leading to vasorelaxation, and inhibition of the ~x- 557 501 · complex; several endothelin-fcisio.forrils and receptors have pression of endothelin-1 converting enzyme; however, both 558 29 31 502 been identified in a variety o'Jktj~shes, including neuronal, receptors have been found to elicit vasoconstriction. - 559 503 renal, ·and vascular tissues.,_ Inte~estingly, although some ETA receptors have been shown to be functionally con- 560 504 data show a correlation betwe~nrihdothelin-1 levels and hy- pled to the Gq111 protein leading to PLC-(3 activation; Gs 561 4 505 pertension, these data are not conclusive. Overall, endothe- linked leading to increased cAMP and also to Gi, thus in-06 562 506 30 32 34 lin-1 signaling in the vasculature is essentially the same as hibiting adenylate cyclase. • - Activation of G011 1, end- 563 507 2 564 for angiotensin II signaling. A brief overview will be dis- ing in IP3 and DAG cleavage, can stimulate Ca + release 508 cussed, but the reader is directed to angiotensin II-GPCR intra.,. and extracellularly, as described for angiotensin · II · 565 509 coupled signaling pathways for a .more iri depth explana- signaling. In addition to second messenger generation, en- 566 510 tion. Differing pathways regarding to endothelin-1 signal- dothelin-1 has also been demonstrated to activate Ca2+ 567 511 ing itself will be discussed further. channels on the plasma. membrane and stimulate Ca2+ 568 512 569 513 570

REV 5.0 DTD ■ JASH136_proof ■ 29 October 2008 ■ 3:57 am ■ ce 6 B.M. Wynne et al. I Journal of the American Society of Hypertension ■ ( ■ ) (2008) 1-13

4 35 2 2 571 flux from the extracellular space. • There is speculation voltage-operated Ca + (VOC), receptor-operated Ca + 628 572 that ETA receptor activation can activate the Na+/H+ ex­ (ROC), store-operated Ca2+ (SOC) channels and a Ca2+­ 629 573 changer and activate the enzymes phospholipase D generat­ permeable nonselective cation channels (NSCC) modulate 630 4 2 574 ing DAG and phospholipase A2 releasing arachidonic acid. most Ca + mobilization within the cell. We will discuss 631 575 ETB receptors in rat carotid were shown to mediate vasore­ these particular vascular smooth muscle Ca2+ entry chan­ 632 57601 laxation via a cGMP-NO pathway, the production of vaso­ nels further in the following ~ection. 633 577 dilator cyclooxygenase products and the activation of 634 578 635 voltage-activated K+ channels. Recently, endothelin-1 Voltage-Operated Ca2 + Channels 579os was shown to inhibit NADPH oxidase activity and superox- 636 580 36 ide generation- via ETB 1 -receptors. The details of these VOC channels represent ·a major route by which Caz+ 637 31 43 581 mechanisms are still poorly understood. · enters vascular s111dotp. cells. VOC channel function is 638 582 Both agonists, angiotensin II and endothelin-1, mediate regulated by merrib'tarie potential such that hyperpolariza- 639 583 vasoconstriction via an increase in cytosolic Ca2+ although tion closes them -arid 4epolarization opens them; the latter 640 584 mechanisms of this process are different between the two leading to vasoionstriction.1 641 585 agonists. Endothelin-1 increases intracellular Ca2+ by stim- The majoriiy ·of- .. agonist-induced Ca2+ influx probably 642 586 ulating influx through Ca2+ channels and is known to pro- occurs thro,bgh ·L;-type VOC channels. Dihydropyridine- 643 587 duce ,a much more sustained contraction by the vascular sensitive, L'~.tyP~. VOC channels are regulated by vasocon- 644 588 smooth muscle cells, whereas angiotensin II elicits a potent stnctor~,;ffrat _a,je.-known to activate the PKC pathway. Vaso- 645 589 biphasic response that is generated primarily by mobiliza- dilators\.have also been shown to inactivate these channels, 646 2 1 4 590 tion of Ca + from an intracellular store. Consequently, but ~ht:9~~h--cAMP production. .4 These gated channels 647 591 the actions of angiotensin II actions are relatively rapid in alsQ;;}i,ejolarize in response to membrane stretch, lending 648 37 1 592. comparison. -4 suppott·to the hypothesis that these channels are important 649 593 650' Given that both of these agonists mediate downstream sig- for1,tp.e myogenic response and vascular tone. Inhibition of 594 naling pathways via the-same second messengers, it is easy to ,(yocl. can occur primarily because of increasing intracellu­ 651 595 2 652 appreciate how cross-talk occurs between them. Both endo- · .;;: ·1aE5,ga + concentration and activation of cGMP-dependent 1 2 4 596 thelin-1 and angiotensin II signal through distinct Gq-coupled '' /priiein kinase. • .4 ·653 597 GPCRs leading to phosphoinositide production and MAP~ :~;;,, .. l'Endothelin-1 has been shown to activate the VOC chan­ 654 598 activation. However, the mechanism by which this occ~s¾;,;__ "nel in smooth muscle ~ells from porcine coronary arteries, 655 2 599 varies at some points. It has been demonstrated that endotpe- -~it;_~ and has also been shown to augment Ca + channel currents 656 -600 lin-1 and angiotensin II both stimulate this pathway vit1Ras-•i(,, in the smooth muscle cell membrane of guinea pig po_rtal 657 601 Raf; yet,. angiotensin II produces phosphorylation of E!gJG(} vein. In a model of cultured thoracic aorta vascular smooth 658 602 2, SAPK/JNK, an~ p38MAPK via c-Src-depend.¢-fit,p"aut:·· muscle cells, Kawanabe et al demonstrated that endothelin- 659 603 ways. In contrast, endothelin~ 1 was shown -tthJndQ·te 1 produces a sustained increase in intracellular Ca2+ levels 660 41 45 604 MAPK phosphorylation through c-Src,in.c;lyp~dent from VOC channels, as well as others. • 661 42 2 605 pathways...... Although endothelin-1 increases intracellular Ca + by 662 606 stimulating influx through Ca2+ channels, the role of 663 664 607 Calcium Mobilizl;ltiori - VOC channels is limited, and current data suggest that 608 other channels are more important; namely the NSCC, as 665 2 4 46 609 Given that Ca + is the trigger for sm9ot1r~Jik:le contrac­ evidenced by experiments . using nifedipine. 1. Even 666 610 tion, it is necessary to understand the ~echa~sms by which though described more than. 60 years ago, the exact mech.: 667 611 intracellular Ca2+ levels increase. Renee/this, is the focus anisrils of action of this peptide are not completely 668 612 of much research, and the subject:,;bf the·::remainder of this understood. 669 613 review. In vivo, arterial smooth"'µi.iis.G:le,,iells exist with an Angiotensin II-GPCR stimulation leading to PLC is· an 670 614 average intracellular Ca2+ concentrat,t9n that is several or­ accepted ·mechanism acknowledged to enhance the L-type 671 615 ders of magnitude lower than t(~{ftti:he extracellular fluid. Ca2+ channel. Interestingly, intracellular angiotensin II 672 616 Intracellular Ca2+ concentra"fi0n.l~~els do not exist homo­ has been demonstrated to stimulate VOC channels simi­ 673 617 genously throughout the c~,U,. b~i]¢xhibit dynamic changes larly, but yet independently of extracellular ligand binding 674 618 2 675 . in the cell, temporally and spa'ti~ly, because of Ca + flux. of angiotensin II to the AT 1 receptor, possibly through the 47 8 619 The changes seen in Ca 2+ flux are a component of several actions of PLC, PKC, and/or tyrosine kinase. .4 676 620 different events, which are all dependent on intrace-llular ul­ As shown in the smooth muscle cells of the anococcy­ 677 621 tra-structure as well as the spatial relationship of ion pumps geus, there are two v~ry distinct mechanisms for inward 678 622 and channels in the plasma membrane.4 Ca2+ currents and Ca2+ store depletion.49 The increase in 679 623 Ca2+ entry through channels in the plasma membrane Ca2+ concentration· after the activation of VOC channels 680 624 and release from the SR ~e the major sources of intracel­ can also lead to a phenomenon known as Ca2+ induced 681 2 1 4 15 2 2 2 625 lular Ca +. • • Of the- different Ca + entry channels the Ca + release, where the increase in Ca + concentration 682 626 683 627 684

RRV 5.0 DID ■ JASH136 proof· ■ 29 October 2008 ■ 3:57 am ■ ce B.M. Wynne et a_l. I Journal of the American Society of Hypertension ■ ( ■ ) (2008) 1-13 7

685 leads to Ca2+ release from intracellular stores. The RyR on proliferation and apoptosis.4 SERCA pumps, which are 742 686 the SR mediates this process, as they are activated by found on the membrane of the sarcoplasmic reticulum, func­ 743 687 Ca2+.so This produces a membrane depolarization that.pro­ tion by pumping the Ca2+ ions back into the.SR after a con­ 744 2 2 14 15 688 motes further VOC channel opening and Ca + entry. tractile event replenishing Ca + concentrations. • SERCA 745 689 pumps can be inhibited by cyclopiazonic acid or thapsigar- · 746 690 gin, which deplete intracellular Ca2+ stores and thereby ac­ 747 Receptor-Operated Ca2 + Channels 66 67 2 691 tivate SOC channels. • Ca + entering through the SOC 748 692 Only 20 years ago, the exact mechanism of Ca2+ channel channels can then be pumped into the stores, replenishing 749 693 activation by extracellular ligands, such as hormones and them. According to functional and pharrnacologic experi- 750· 694 neurotransmitters, was largely unknown. In addition, there · ments, SOC channel entry elicited by specific inhibitors of 751 695 was still uncertainty concerning the mechanism by which SERCA pumps induced a sustained elevation of intracellular 752 2 696 ca2+· entry was mediated until a model, proposed by Put- Ca2+, which is also d'e.pendent on extracellular Ca + entry 753 697 ney et al, detailing the mechanism by which activation of and is attribute9-f

799 Nonselective Cation Channels As described previously, Pyk2 is a Ca2+ -sensitive nonre- 856 I 800 ceptor protein tyrosine kinase that is known to associate 857 801 The NSCC, so called because it is "nonselective," being With focal 8idhesion sites and is a downstream effector of 858 equally permeable to monovalent cations, such as Na+ and 7 15 80 81 802 • • • . 859 25 28 angiotensin II and endothelin-1 OPCR stimulation. 803 K+ in the extra- and intracellular compartments. • This Pyk2 phosphorylation and activation has been postulated in 860 804 channel has been revealed to be activated by stimulation several different pathways, including ERKl/2 signaling in 861 805 of native ETA receptors in vascular smooth muscle, and cardiomyocyt~s, p38MAPK in mesangial cells and recently 862 806 the majority of endothelin-1 response being mediated by in RhoA/Rho kinase activation via PDZ-RhoGEF vascular 863 1 6 75 NSCCs; namely, NSCC-1 and NSCC-2.28.4 .4 • It was 15 80 82 83 807 smooth muscle cell migration. • • • It has even been 864 808 previously shown, pharmacologically and with electrophys­ suggested that tyro~ine kinase pathways such as these 865 809 iology, that the response to endothelin-1 is dose-dependent. may even be impoyfanJ- in angiotensin II signaling leading 866 810 More specifically, with lower concentrations of endothelin- 4 867 2 to vasoconstrictiori:':~ }fhese alternative signaling pathways 811 1, Ca + entry via IP3 formation from intracellular stores oc­ in angiotensin :r;r,,sfgtiaj.1iig should be a new directive for in- 868 812 curs; however, with higher endcithelin-1 concentrations vestigation int&.incre~sed vascular smooth muscle reactiv- 869 2 813 both _the former in addition to extracellular Ca + entry oc­ ·ity in condi!ioµl;sueh as hypertension. This was a focus 870 curs as well.2s,2s,3s,76,77 814 of a recent aiticie'from Oiachini et· al. 85 It is known that in- 871 815 The molecular mechanisms of NSCC activation are still creased va;~~~~_.,reactivity to contractile stimuli is present 872 not entirely clear. Recent evidence in rabbit internal carotid 816 in vessclf'1ir,Qm'l,DOCA-salt hypertensive mice. Using this QIO 873 817 artery shows that endothelin-1 may induce an intracellular model :Wit,l(~"Iiharmacological approach, mechanistic stud- 874 818 response after OPCR activation via Pyk2 and the ETA ies wer~,performed to determine the role that Pyk2 plays in 875 819 receptor. Data from the same laboratory indicate that phos­ the,·typ.erreactivity exhibited in DOCA-salt hypertension. 876 2 820 phoinositide-3 kinase also regulates the activation of Ca + In mes'en~eric arteries and aorta, Pyk2 inhibition attenuated 877 821 entry after endothelin-1 GPCR activation and stimulation of Jhe,)ncreased vascular constriction to phenylephrine and 878·; · 46 78 822 NSCC-2. · . Western blot data showed increased Pyk2 and phospho- 879 823 •· Py~·: in vessels from DOCA-salt treated mice vs. sham, 880 824 80 85 Hot Topics in Vascular Smooth Muscle Signaling .. .. ··th~s· confirmi~g a role for Pyk2 in_h~peitension. • 881 825 . _,:: ·.. ,,.,,. -~./A recent discovery of a transcnpt1on factor named the 882 826. Within just the past year, exciting n~w insights into v.is,.'ll;~ . ·repressor element !-silencing transcriptional (REST) 883 1 827 cular smooth muscle signaling have been discovered._.:We ·'.\::, factor, was shown to have a consensus sequence, KCNN4, 884 828 will discuss several recent advancements in the fie\('.that~;i: which is known to encode for the intermediate conductance 885 829 have been especially stimulating. ~l;.,. j- Ca2+ -activated K+ channels (IKcJ. These channels are of 886 830 As described previously, OPCR signaling can eLi;~it;§ffifl.:" the utmost importance in small resistance vessels where 887

831 'ulatory or inhibitory signals in the cell. With reg .g; .'?.! a primary vasodilating agent is endothelium-derived hyper- 888 832 cular smooth muscle, the Gq1G 11 · G protein polarizing factor. En~othelium-derived hyperpolarizing fac- 889 833 accepted as producing contraction via the C. tor not only remains elusive as a factor, but the mechanism by 890 834 PLC pathway; however, Ca2+ -independe11-t-s1:o "Jmisms, which it produces vasodilation is only speculation. One such· 891 835 not only for force maintenance, but for ,¢Gn:~ae.tion have hypothesis is that K+ efflux from endothelial cells via these 892 11 836 emerged. By means of a. novel murine knqf~out model IKca channels in coordination with small-conductance 893 837 for Gq/G 11 and 0 12/013 0 proteins in smoot~ iriuscle cells, Caz+ -activated K+ channels, activates inward rectifier K+ 894 838 Wirth et al elaborated upon the knowi~dge ~f how angio- channels. This would lead to vascular smooth muscle relax- 895

839 • tensin II and endothelin-1 mediate0 .00ritfacfion via G pro- ation. Unpublished results from the laboratory of Webb and 896 840 teins. Using aortic -segments th5:y derrt_pnstrated that in Tastes show for the first time that REST expression is closely 897 841 OiifG 13 knockout mice, angiot.1t..n§lll,,, .U/and endothelin-1 associated with IKca expression in arteries from hypertensive 898 842 were able to elicit a contractio~';'·'~o.wever, both potency animals. Giachini et al demonstrated that REST expression 899 843 and efficacy were affected. In.-:;thi:h~/G 11 knockout mice, was downregulated, whereas IKca channels were overex- 900 844 although endothelin-1 still °];ir-Q,g_µc:ed a severely reduced pressed in arteries obtained from spontaneously hypertensive 901 845 contraction, angiotensin LI:.,.~as :'.t11ot capable of eliciting stroke-prone rats, when compared to their Wistar-Kyoto 902 846 2 any contraction. It was also de'm~hstrated that G1zf0 13 cou- controls. Protein levels of small-conductance Ca +-activated 903 847 pled G proteins activate the RhoA/Rho kinase signaling K+ channels and IKca were also assessed, and alterations 904 848 pathway via interaction of RhoGEFs; in particular, were observed in mesenteric arteries from hypertensive 905 84Q9 LARG. Using a LARO murine knockout model, they animals. It seems as though REST may be a negatively mod- 906 850 showed that LARO is necessary for full contraction with ulatory mechanism, which controls the levels of IKca in the 907 851 angiotensin II and endothelin-1, as demonstrated by the se- spontaneously hypertensive stroke-prone rat vasculature. 908 852 verely restriction constriction elicited in aortic segments As mentioned previously, the STIM molecule was shown 909 10 79 853 from LARG knockout mice. • to play an essential role for SOCE and conductance through· 910 854 911 855 912

n-r,,u ,, n rvrn - TA ,;:u.1 -:ii; nrnnf - ')Q nrtnhe,r ?OOR • ~-'i7 am ■ ce B.M. Wynne et al. I Journal of the American Society of Hypertension ■ ( ■ ) (2008) 1-13 9

913 CRAC channels.71 The recently discovered subunit of the vascular smooth muscle. The contractile apparatus of vas- 970 914 CRAC channel pore,. termed Orail, is essential for CRAC cular smooth muscle, actin, and myosin, can be activated 971 62 72 86 87 2 2 915 channel activation. • • • However, the exact mechanism in a Ca +-dependent and Ca +-independent manner. Via Ii- 972 916 for STIM/Orail signaling is not yet known. STIM can un- gand binding to plasma membrane GPCRs, second messen- 973 917 dergo phosphorylation at specific serine/threonine sites, as gers are generated and induce the release of Ca2+ through 974 918 well as undergo N-linked glycosylation.88 Sobolff et al de- channels located on the plasma membrane or on the SR 975 919 termined that although STIMl is expressed at the cell sur- producing a rapid and transient increase in intracellular 976 920 face and within the endoplasmic reticulum (ER), the Ca2+. Channels discussed in this review are activated 977 921 STIM2 protein is expressed only intracellularly, which through ligand binding, store depletion, and membrane de- 978 922 likely reflects an ER-retention signal that is present in polarization. In the .form of a Ca2+-calmodulin complex, 979 8 90 923 STIM2 but riot STIMl. 6- It was apparent that suppressed subsequent activation of MLCK occurs, inducing contrac- 980 924 STIMl expression,' but not STIM2, was able to prevent tion via actin-mybsin; cross-bridges. Contraction also oc- 981 925 SOCE and eliminate the store-dependent activation of curs, Ca2+ indetfertde~tly, through the activation of the . 982 71 73 926 CRAC channels. • Thus the function of STIMl was pas- RhoNRho kinase pat:Jlway leading to MLCP inactivation, 983 927 tulated to act as a Ca2+ sensor in the ER.72 and the maintenartee of contraction. 984 928 On a final note, the question of "blame" for hyperten- Given the vari~d mechanisms for smooth muscle contrac- 985 929 sion is argued between two schools of thought: cardiovas- tion, one ca-n.i~!l,g1ne the wide range of areas where dereg- 986 930 cular and renal physiologists; the former believing that ulation .can:,'_q~_cur, leading to increased BP or hypertension. 987 931 hypertension is due to increased vascular resistance and On a m~l~cuiar' level, alterations in various cellular signal- 988 932 an overail hyperreactivity of vascular tissue, the latter be- .ing ~dEl;l._pbnents, can lead to over stimulation, causing an 989 933 lieving that the kidney has the final say in BP control. _Until incte.~sJtl and maintained vasoconstriction, decreased re- 990 934 recently, renal physiologists everywhere cited articles by lax~ti~fi%_and, consequently, elevati~n of systemic BP. 991 935 Guyton, who was the first to clearly demonstrate the phe- .it·~t~. 992,. 936 nomenon of "pressure natriuresis" and argue for the central :.t.Uncl!ted reference 993 937 994 role for the kidney in BP control, and Coffman, who uses •, "·y-"·:.::_.0 /l 938 a model of AT lA receptor knockout to illustrate the neces-i'' / ·io. 995 91 92 939 sity of the RAS in hyperterision. • These elegant studie~•,;,·:-:,{, .. ,/' 996 '940 used the AT 1 receptor knockout mice and kidney era_~~;,;,... 'tJncited Figures 997 941 transplantation to show that the AT lA receptor is c~§}al ... ,,,b1, 998 942 to basal BP regulation, as well as hypertension, an(lf'thah, Figure 1-4 999 943 AT lA rec_eptors in the kidney are paramount in hypert6i\§Jon.) 1000 944 and the renal response to hypertension. Howey~t, .,th~it References 1001 945 studies also demonstrated that BP regulation byfA'.fi.A re­ 1.002 946 ceptors in nonrenal tissues (ie, vasculature) were,·als~·a,liha­ 1. Jackson WF. Ion channels and vascular tone. Hyperten­ 1003 93 94 947 jor contributor to systemic BP. · An artj'.c.ley'\·ecently sion 2000;35:l73_;8. 1004 948 published in PNAS by Michael et al comp,J.erltea,.ts these 2. Hilgers RH, Webb RC. Molecular aspects of arterial 1005 949 studies, using a knocking mutant of the ,iGI\1-P"'q~pendent smooth muscle contraction: focus on Rho. Exp Biol 1006 950 protein kinase, PKGfo. PKGI is expressed0'1~fvascular and Med (Maywood) 2005;230:829-35. 1007 951 smooth muscle cells and has been show,1f'td•,;11igulate vascu­ 3. Woodrum DA, Brophy CM. The paradox of smooth 1008 952 lar relaxation via endothelial-derived ~2 an1~ther nitrova­ muscle physiology. Mol Cell Endocrinol 2001;177: · 1009 953 sodilators. The most remarkable .dat;afrcittr··tliis study show 135-43. 1010 954 that the LZM-PKGia mutant mi4'.f exlfipit elevated BPs, 4. Wamhoff BR, Bo\\'.les DK, Owens GK. Excitation­ 1011 955 even in the presence of normal ,,terta.:1J,.uiicti6n and normal transcription coupling in arterial smooth muscle. Circ 1012 956 renal salt handling, suggesting i~=~•::iwportant mechanism Res 2006;98:868-78. 1013 957 for vascular smooth muscle ii'f-tgfnormal and pathophysi­ 5. Higuchi S, Ohtsu H, Suzuki H, Shirai H, Frank GD, 1014 9 958 ologic control of BP. ~ Ovefurl~jfiese studies reiterate our Eguchi S. Angiotensin II signal transduction through 1015 959 awareness of the comple~ity cif hypertension, and the the ATl receptor; novel insights into mechan_isms and 1016 960 necessity for systemic integr1i.tiori~ pathophysiology. Clin Sci (Land) 2007;112:417-28. 1017 961 6. Ohtsu H, Suzuki H, Nakashima H, Dhobale S, 1018 962 1019 Conclusion Frank .GD, Motley ED, et al. Angiotensin II signal 963 transduction through small OTP-binding proteins: 1020 964 Arterial vascular smooth muscle cells constitute the ma­ mechanism and significance in vascular smooth muscle 1021 965 jority of the arterial wall, playing a foremost tole in vascu~ cells. Hypertension 2006;48:534-40. 1022 966 lar resistance and blood flow. Angiotensin II and 7. Touyz RM, Berry C, Recent advances in angiotensin II 1023 967 endothelin-1 are potent agonists inducing contraction of signaling. Braz J Med Biol Res 2002;35:1001-15. 1024 968 1025 969 1026

REV 5.0 DTD ■ JASH136 oroof ■ 29. October 2008 ■ 3:57 am ii ce 10 . B.M. Wynne et al. I Journal of the American Society of Hypertension ■ ( ■ ) (2008) 1-13

1027 8. Matsubara H. Pathophysiological role of angiotensin II 22. Chitaley K, Weber D, Webb RC. RhoA/Rho-kinase, 1084 1028 type 2 receptor in cardiovascular and renal diseases. vascular changes, and hypertension. Curr Hypertens 1085 1029 Circ Res 1998;83:1182-91. Rep 2001;3°:139-44. 1086 1030 9. Harris DM, Cohn HI, Pesant S, Zhou RH, Eckhart AD. 23. Schmidt A, Hall.A. Guanine nucleotide exchange fac­ 1087 1031 Vascular smooth muscle G(q) signaling is involved in tors for Rho GTPases: turning on the switch. Genes 1088 1032 high blood pressure in both induced renal and genetic Dev 2002;16:1587-609 .. 1089 1033 vascular smooth muscle-derived models of hypertension. 24. Uehata M, Ishizaki T, Satoh H, Ono T, Kawahara T, 1090 1034 Am J Physiol Heart Circ Physiol 2007;293:H3072-9. Morishita T, et al. Calcium sensitization of smooth 1091 1035 10. Wirth A, Benyo Z, Lukasova M, Leutgeb B, muscle mediated by a Rho-associated protein kinase 1092 1036 Wettschureck N, Gorbey S, et al. Gl2-G13-LARG­ in hypertension.,Nature 1997;389:990-4. 1093 1037 mediated signaling in vascular smooth muscle is re­ 25. Miwa S, Iwam~ro Y, Zhang XF, Inoki T, Okamoto Y, 1094 1038 quired for salt-induced hypertension. Nat Med 2008; . Okazawa M, er-at:.;.~a2+ entry channels in rat thoracic 1095 1039 14:64-8. aortic smoqth 'tniscle cells activated by endothelin-1. 1096 1040 11. Ohtsu H, Higuchi S, Shirai H, Eguchi K, Suzuki H, Jpn J Phanµacol t999;80:281~8. 1097 1041 Hinoki A, et al. Central role of Gq in the hypertrophic 26. Yanagis~_w..i~:'1-1~•:' 1Inoue A, Ishikawa T, Kasuya Y, 1098 1042 signal transduction of angiotensin II in vascular smooth KimunfS, Kajnagaye S, et al. Primary structure, syn­ 1099 1043 muscle cells. Endocrinology 2008;149:3569-75. thesis, ~Q bi¢1ogical activity of rat endothelin, an en­ 1100 1044 12. Woodsome TP, Polzin A, Kitazawa K, Eto M, dotqeli~m:derived vasoconstrictor peptide. Proc Natl I 101 1045 Kitazawa T. Agonist- and depolarization-induced sig­ M~st,Sti tJ s A 1988;85:6964-7. II02 1046 nals for myosin light chain phosphorylation and -force 27. }flf~gf~rawa M, Kurihara H, Kimura S, Goto K, 1103 1047 generation of cultured vascular smooth muscle cells. .,i1,1a,s'aki T. A novel peptide vasoconstrictor, endothelin,. I 104 1048 J Cell Sci 2006; 119: 1769-80. is:''produced by vascular endothelium and modulates I 105 1049 13. Del Valle-Rodriguez A, Calderon E, Ruiz M, Ordonez A, .,.-, ,·:: §mooth muscle Ca2+ channels. J Hypertens Suppl ll06 1050 Lopez-Barneo J, Urena J. Metabotropic ca<2+) channe­ 'f988;6:S188-91. Il07 1051 l-induced ca<2+) release and ATP-dependent facilitation '28:,,Jv.liwa S, Kawanabe Y, Okamoto Y, Masaki T. Ca2+ en­ 1108 1052 of arterial myocyte contraction. Proc Natl Acad Sci US A_.. _·,· try channels involved in endothelin-1-induced contrac­ Il09 1053 2006;103:4316-21. ,- tions of vascular smooth muscle cells. J Smooth Muscle IIIO 1054 14. Urena J, del Valle-Rodriguez A, Lopez-Barneo J. Me::; Res 2005;41:61-75. 1111 1055 tabotropic Ca2+ channel-induced caleium release,. in · '·.', 29. _Mohacsi A, Magyar J, Tamas B, Nanasi PP. Effects of Ill2 1056 . vascular smooth muscle. Cell Calcium 2091)i.2: endothelins on cardiac and vascular cells: new thera­ II 13 1057 513-20. . . peutic target for the future? Curr Vase Pharmacol I 114 1058 15. Bouallegue A, Daou GB, Srivastava AK. Nitpc 9xid~ 2004;2:53-63. Ill5 1059 attenuates endothelin-1-induced activation of;,a~kl/42, 30, Neylon CB. Vascular biology of endothelin signal Ill6 1060 PKB, and Pyk2 in vascular smooth mu~i*~> ~2lts:f."by transduction. Clin Exp Pharmacol Physiol 1999;26: 1117 1061 a cGMP-dependent pathway. Am J Physi~l IJbarf:-circ 149-53. 1118 1062 Physiol 2007;293:H2072-9. .,,,.':••l,f_", .. 31. Tirapelli CR, Casolari DA, Yogi. A, Montezano AC, 1119 1063 16. Lesh RE, Nixon GF, Fleischer S, Aire}f"JA/t8:G)µilyo AP, Tastes RC, Legros E, et al. Functional characterization 1120 .1064 Somlyo AV. Localization of ryanodln:&.,fJeceptors in and expression of endothelin receptors in rat carotid I 121 1065 smooth muscle. Circ Res 1998;82,;::F15~s,f'· artery: involvement of nitric oxide, a vasodilator 1122 1066 17. Hirano K. Current -topics in _the r~gulato.fu' mechanism prostanoid and the opening of K + channels in 1123 1067 underlying the Ca2+ sensitizatt~n of'

1141 35. Smith L, Payne . JA, Sedeek MH, Granger JP, 49: Wayman CP, McFadzean I, Gibson A, Tucker JF. Two 1198 1142 Khalil RA. Endothelin-induced increases in Ca2+ entry distinct membrane currents activated by cyclopiazonic 1199 1143 mechanisms of vascular contraction are enhanced dur­ acid-induced calcium store depletion in single smooth 1200 1144 ing high-salt diet. Hypertension 2003;41:787-93. muscle cells of the mouse anococcygeus. Br J Pharma­ 1201 1145 36. Dammanahalli JK, Sun Z. Endothelin-1 Inhibits col 1996;117:56.6-72. 1202 1146 NADPH oxidase activity in human abdominal aortic 50. Zhang AY, Li PL. Vascular physiol~gy of a Ca2+ mobi­ 1203 1147 endothelial c~lls: a novel function of ETB 1 receptors. lizing second messenger - cyclic ADP-ribose. J Cell 1204 1148 Endocrinology 2008;149:4979-87. Mal Med 2006; 10:407-22. 1205 1149 37. Douglas JT, Curiel DT. Strategies to accomplish tar­ 51. Putney JW Jr. A model for receptor-regulated calcium 1206 1150 geted gene delivery to muscle cells employing tro­ entry. Cell Calcl:µm 1986;7:l-12. 1207 1151 1208 pism-modified adenoviral vectors. Neuromuscul 52. Benham CD, ij€ss,1P, Tsien RW. Two types of calcium 1152 Disord 1997;7:284-98. channels in sirigle.-: smooth muscle cells from rabbit ear 1209 1153 38. Rubanyi GM, Polakoff MA. Endothelins: molecular bi­ artery stucµea:e~;,,~ftli whole-celt and single-channel 1210 1154 ology, biochemistry, pharmacology, physiology, and recordings}~irc ijes 1987;61:Il0-6. 1211 1155 pathophysiology. Pharmacol Rev 1994;46:325-415. 53. Morris APt~tf~Hacher DV, Irvine RF, Petersen OH. Syn- . 1212 1156 39. Komuro T, Miwa S, Zhang XF, Minowa T, Enoki T, ergism(of i~&~tol tr£sphosphate and tetrakisphosphate 1213 1157 Kobayashi S, et al. Physiological role of in acti~a~ipg)Ca2+-dependent K+ channels. Nature 1214 1158 2 1215 Ca +-permeable nonselective ~ation channel in endo-. 198'il':3'iQ:653-5.r .. ··..::.· 1159 thelin-1-induced contraction of rabbit aorta. J Cardio­ 54. Kunp ~. Gardner P. Ion channels activated by inositol 1216 1160 vasc Pharmacol 1997;30:504-9 . h4,•5~tri:sphosphate in plasma membrane of human 1217 . 1161 40. Dostal DE, Murahashi T, Peach MJ. Regulation of cy­ /f:.!1inphocytes. Nature 1987;326:301-4. . 1218 1162 tosolic calcium by angiotensins in vascular smooth 55. Saleh SN, Albert AP, Peppiatt CM, Large WA. Angio­ 1219 1163 muscle. Hypertension 1990;15:815-22. ·.,... .tensin II activates two cation conductances with distinct 1220 1164 41. Kawanabe Y, Hashimoto N, Masaki T. Characterization · ·11RPC1 and TRPC6 channel properties in rabbit rne~en- 1221 1165 of Ca2+ channels involved in endothelin-1-induced . ··;._•·J.eric artery rnyocytes. J Physiol 2006;577:479-95. 1222 1166 contraction of rabbit basilar artery. J Cardiovasc Phar­ -/"5~ •. -Tai K, Hamaide MC, Debaix H, Gailly P, Wibo M, 1223 1167 1224 macol 2002;40:438-47. ,.;· ,,,.,. .,~ .. ~~,t.:- Morel N. Agonist-evoked calcium entry in vascular 1168 42. Yogi A, Callera GE, Montezano AC, Aranha ;\$~:, .. smooth muscle cells requires IP3 receptor-mediated ac- 1225 1169 Tastes RC, Schiffrin EL, et al. Endothelin-1, but 'hot'•\'.> tivation of TRJ,>Cl. Eur J Pharmacol 2008;58~:135-47. 1226 . 4fi:i} 1170 Ang II, activates MAP kinases through c-Src indepQi\;7;-"m cation channel with TRPC3 and TRPC7 properties in 1229 f ,,· ... 1173 43. Bolton TB. Mechanisms of action of trans~~rs ~fld rabbit coronary artery myocytes. J Physiol 2007;580: 1230 1174 other substances on smooth muscle. f~§io1t"~R.ev 755-64. 1231 1175 1979;59:606-118. l r~·,,..,". 58. Smyth JT, Dehaven WI, Jones BF, Mercer JC, 1232 . '""'iii" 1176 44. Hughes AD. Calcium channels in v ·1ac~,smooth Trebak M, Vazquez G, et al. Emerging perspectives 1233 2 1177 muscle cells. J Vase Res 1995;32:35~;# ,tr :,.,,.:· in store-operated Ca + entry: roles of Orai, Stirn and 1234 1178 45. Goto K, Kasuya Y, Matsuki N, Takuwttni2e1 2003;63:808-1:3. 62. Leung FP, Yung LM, Yao X, Laher I, l{uang Y. Store-­ 1243 1187 47. Eto K, Ohya Y, Nakam:l.,.\ra Y,j.\be I, Iida M. Intracellu­ operated calcium entry in. vascular smooth muscle: Br J 1244 1188 lar angiotensin II stim'ul~tef voltage-operated ca<2+) Pharmacoi 2008;153:846-57. 1245 1189 channels in arterial myocytes. Hypertension 2002;39: 63. Putney JW Jr. Capacitative calcium entry revisited. Cell 1_246 1190 474-8. Calcium i990;11:611-24. 1247 1191 48. Ohya Y, Sperelakis N. Involvement of a GTP-binding 64. Putney JW Jr. Receptor-regulated calcium entry. Phar- 1248 1192 protein in stimulating action of angiotensin II on cal­ macol Ther 1990;48:427-34. '1249 1193 cium channels in vascular smooth muscle cells. Circ 65. Putney JW Jr. The integration of receptor-regulated in- 1250 1194 Res 1991;68:763-71. tracellular calcium.release and calcium entry across the 1251 1195 1252 1196 1253 1197 1254

REV 5.0 DTD ■ JASH136_proof ■ 29 October 2008 ■ 3:57 am ■ ce 12 B.M. Wynne et al. I Journal of the American Society of Hypertension ■ ( ■ ) (2008) 1-13

1255 plasma membrane. Curr Top Cell Regul 1990;31: activation of v.oltage-independent Ca2+ channels and 1312 1256 111-27. vasoconstriction. Biochem Pharmacol 2004;68: 1313 1257 66. Kasuya Y, Ishikawa T, Yanagisawa M, Kimura S, 215-21. 1314 1258 Goto K, Masaki T. Mechanism of contraction to endo- 79. Schoner W. ·salt abuse: the path to hypertension. Nat 1315 1259 thelin in isolated porcine coronary artery. Arn J Physiol Med 2008;14:16-7. 1316 1260 1989;257:H1828-:--35. 80. Fernarida RC, Giachini FSC, Lima VV, Carniero ZN, 1317 1261 67. Takemura H, Thastrup 0, Putney JW k Calcium efflux Carvalho MHC, Fortes ZB, et al. Pyk2 mediates in- 1318 i262 across the plasma membrane of rat parotid acinar cells creased adrenergic contractile responses in arteries 1319 1263 is unaffected by receptor activation or by the micro- from DOCA-salt mice. J Am Soc Hypertens 2008;. Qtt1320 1264 somal calcium ATPase inhibitor, thapsigargin. Cell Forthcoming. 1321 1265 Calcium 1990;11:11-7. . 81. Yin G, Yan C,):lerl,< BC. Angiotensin II signaling path- 1322 1266 68. Putney JW Jr, Broad LM, Braun FJ, Lievremont JP, ways mediated'--b.y.;tyrosine kinases. Int J Biocµem Cell 1323 1267 Bird GS. Mechanisms of capacitative calcium entry. Biol 2003;3•5:78-0..,..3. · 1324 1268 J Cell Sci 2001; 114:2223-9. 82. Bouallegue.. ~ A, D~ou GB, Srivastava AK. Endotheli- 1325 1269 69. Bouallegue A, Yamaguchi N. Nitric oxide inhibits the n-l-indu9~d "si:griiling pathways in vascular smooth 1326 1270 bradykinin B2 receptor-mediated adrenomedullary cate- muscle.icells>Curr Vase Pharmacol 2007;5:45-52. 1327 1271 cholamine release but has no effect on adrenal blood 83. Schallet~,MD: Calcium-dependent Pyk2 activation: 1328 1272 flow response in vivo. J Pharmacol Sci 2005;98:151-60. a role·fox:.·~·;1modulin? Biochem J 2008;410:e3-4. 1329 1 1273 70. Venkatachalam K, van Rossum DB, Patterson RL, 84, Betls-_,;1,ic'. Angiotensin II signal transduction in vascu- 1330 1274 ~a HT, Gill DL. The cellular and molecular basis of lat;r§m.doth muscle: pathways activated by specific ty- 133 l 1275 store-operated calcium entry. Nat Cell Biol 2002;4: :tfasjie kinases. J Arn Soc Nephrol 1999;11(10 Suppl): 1332 1276 E263-72. S'62~8. 1333 1277 71. Roos J, DiGregorio PJ, Yeromin · AV, Ohlsen K, .!~:5:;;.':il>::~iiachini Frc CF, Lima VV, Carneiro ZN, 1334 1278 Lioudyno M, Zhang S, et al. STIMl, an essential and _;::· G;~alho MHC, Fortes ZB, Webb RC, et al. Pyk2 me- 1335 1279 conserved component of store-operated Ca2+ channel :, ·,;_,,~";.,.,~iates increased adrenergic contractile responses in ar- 1336 1280 function. J Cell Biol 2005;169:435-45. Jr ,1,:' ,.:teries from DOCA-salt mice. J Arn Soc Hypertens 1337 1281 72. Zhang SL, Yu Y, Roos J, Kozak JA, Deerinck Tl/\'-.:;~;, . ..:,l 2008. Forthcoming. · Q121338 1282 Eliisman MH, et al. STIMl is a Ca2+ sensor that a~,!Jf.. ,,,.. "'«6. Soboloff J, Sp.assova MA, Tang XD, Hewavitharana T, l339 1283 v_ates CRAC channels and migrates from the .Ca2+ st9re ''~,,. Xu W, Gill DL. Orail and STIM reconstitute store-op- 1340 1284 to the plasma membrane. Nature 2005;437:902-?,';·'· ·· .,:. erated calcium channel function. J Biol Chem 2006; 1341 1285 73. Liou J, Kim ML, Heo WD, Jones JT, MyerSC,}Y\',,} 281:20661-5. 1342 1286 Ferrell JE Jr, et al. STIM is a Ca2+ sensor,,essentfin·' 87. Spassova MA, Soboloff J, He LP, Xu W, Dziadek MA, 1343 1287 for ca2+-store-depletion-triggered Ca2+ inft~.--· C_µrr Gill DL. STIMl has a plasma membrane role in the ac- 1344 . 1288 Biol 2005;15:1235-41. ('\.,·, :. -··· tivation of store-operated ca<2+> channels. Proc Natl 1345 1289 74. Takahashi Y, Watanabe H, Murakami ,.J1,,:Ohba T, Acad Sci US A_2006;103:4040-5. 1346 1290 Radovanovic M, Ono K, et al. Involve~a~,t'df::~ansient 88. Manji SS, Parker NJ, Williams RT, van Stekelenburg L, 1347 1291 receptor potential canonical 1 (TRPCl'') ijfan.,giotensin Pearson RB, Dziadek M, et al. STIM 1 : a novel phos- l 348 1292 II-induced vascular smooth muscle c'en;;J1ypertrophy. phoprotein located at the cell surface. Biochim Biophys 1349 1293 Atherosclerosis 2007;195:287-96./.. -.,:,..,," . Acta 2000;1481:147-55. 1350 1294 75. Chen C, Wagoner PK. Endotheli1:i:jnduc~ a nonselec- 89. Soboloff J, Spassova MA, Dziadek MA, Gill DL. Cal- 1351 1295 tive cation current in vascular,,slrib°'eth" muscle cells. cium signals mediated by STIM and Orai proteins--:-a . 1352 1296 1 Circ Res 1991;69:447-54. t '\ new paradigm in inter-organelle communication. Bio- 1353 1297 · 76. Enoki T, Miwa S, Sakamotqifl~:t~!il,Bwa T, Komuro T, chim Biophys Acta 2006;1763:1161-8. 1354 1298 Kobayashi S, et al. Lon~~lasfri'lg,J1ctivation of cation 90. Soboloff J, Spassova MA, Hewavitharana T, He LP, 1355 1299 current by low concentratfiit,'&"endothelin-1 in mouse Xu W, Johnstone LS, et ~I. STIM2 is an inhibitor of 1356 1 1300 fibroblasts and smooth rrllis~i "b~lls ofrabbit aorta. Br J STIMl-mediated store-operated Ca2 + Entry. Curr 1357 1301 Pharmacol 1995;115:479-85:'lt Biol 2006;16:1465-70. 1358 1302 77. Kasuya Y, · Takuwa y;·,, Yafiagisawa M, Kimura S, 91. Coffman TM, Crowley SD. Kidney in hypertension: 1359 1303 Goto K, Masaki T. Endothelin-1 induces vasocortstric- Guyton redux. Hypertension 2008;51:811-6. 1360 1304 tion through two functionally distinct pathways in por-. 92. Guyton AC. Blood pressure control-special role of the 1361 1305 cine coronary artery: contribution of phosphoinositide kidneys and _body fluids. Science 1991 ;252: 1813-6. 1362 1306 turnover. Biochem Biophys Res Commun 1989;161: 93. Coffman TM, Himmelstein S, Best C, Klotman PE. 1363 1307 1049-55. Post-transplant hypertension in the rat: effects of capto- 1364 1308 78. Kawanabe Y, Hashimoto N, Masaki T. Effects of pril and native nephrectomy. Kidney Int 1989;36: 1365 1309 phosphoinositide 3-kinase on endothelin-1-induced 35-40. 1366 1310 1367 1311 1368

REV 5.0 DTD ■ JASH136_proof ■ 29 October 2008 ■ ~:57 am ■ ce B.M. Wynne et al. I Journal of the American Society of Hypertension ■ ( ■ ) (2008) 1-13 13

1369 94. Crowley SD, Gurley SB, Herrera MJ, Ruiz P, Griffiths R, a defect in vascular function:. Proc Natl Acad Sci U S 1376 1370 Kumar AP, et al. Angiotensin II causes hypertension and A 2008;105:6702-7. 1377 1371 1378 1372 cardiac hypertrophy through its receptors in the kidney. 96. Seshiah PN, Weber DS, Rocic P, Valppu L, 1379 1373 ·Proc Natl Acad Sci US A 2006;103:17985-90. Taniyama Y, Griendling KK. Angiotensin II stimulation 1380 1374 95. Michael SK, Surks HK, Wang Y, Zhu Y, Blanton R, of NAD(P)H oxidase.activity: upstream mediators. Circ 1381 -1375 Jamnongjit M, et al. High blood pressure arising from Res 2002;91:406-13. 1382

REV 5.0 DTD ■ JASH136 proof ■ 29 October 2008 ■ 3:57 am ■ ce Mesenteric arteries from spontaneously hypertensive stroke prone rats·

exhibit an increase in NO-dependent vasorelaxation via iNOS.

1 2 1 2 *B.M. Wynne MS1; *F.R. Giachini PhD • , H. Labazi1, V.V. Lima MS • , F.

1 2 3 Carneiro MS • , Anne M. Dorrance PhD , R. Clinton Webb PhD1, R.C. Tostes

1 PhD ,4

1Medical Colleg·e of Georgia, Department of Physiology, Augusta, GA, USA. 2Pharmacology, Institute of Biomedical Sciences, University of Sao Paulo, Sao Paulo, SP, Brazil; 3Department of Pharmacology and Toxicology, Michigan State University, East · Lansing, Ml, United States _ 4 Department of Pharma·cology, Medical School of Ribeirao Preto, University of Sao Paulo, SP, Brazil

* These authors equally contributed to studies reported in this manuscript. .

Corresponding Author:

Brandi·M. Wynne 1120 15th Street CA 3099 Department of Physiology Georgia Health Sciences University Augusta, GA 30912 phone: 706.721.3547 fax: 706. 72 t; 7299 email: [email protected]

Funding: This study was supported by grants from FAPESP - Fundacao de Amparo a Pesquisa de Sao Paulo to RCT and the National Heart, Lung and Blood Institute at the National Institutes of Health [5R01 HL071138-08] to RCW.

Running title: Increased NO-dependence in SHRSP mesenteric arteries. Key Words: vascular, inducible nitric oxide synthase, hypertension, SHRSP

Abbreviations:

Nitric oxide, NO; mean arterial pressure, MAP; nitric oxide synthase, NOS; inducible nitric oxide synthase, iNOS; neuronal nitric oxide synthase, nNOS;

Wistar-Kyoto, WKY; spontaneously hypertensive stroke-pron~ rats, SHRSP; cyclooxygen.ase, COX; Nw-nitro-L-arginine methyl ester, L-NAME; acetylcholine,

ACh; phenylephrine, Phe; concentration response curve, CRC; reactive nitrogen species, RNS; reactive oxygen species, ROS; ~ndothelial hyperpolarizing factor,

EDHF; calcium, Ca2+; soluble guanylate cyclase, sGC; cyclic guanosine monophosphate, cGMP; nuclear factor kappa-light-chain-enhancer of activated B

cells, NF-KB; phosphoin_ositide 3 kinase, Pl3K; serine, ser; vascular smooth muscle, VSM Abstract

The pathophysiology of hypertension is known to include increased peripheral vascular resistance, inflammatory mediators and endothelial dysfunction. The . endothelium is crucial for the mainter:,ance of vascular tone by releasing several vasoactive substances, including nitric oxide (NO). In this study, we aimed to determine the mechanism for vasorelaxation in the mesenteric arteries of the spontaneou~ly hypertensive stroke-prone rat (SHRSP). We hypothesized that mesenteric arteries of SHRSP up-regulate eNOS, leading to increase NO production to compensate for the endothelial dysfunction. Mesenteric arteries were isolated for use in functional studies or Western blot analysis. NOS and COX inhibition resulted in a decreased relaxation response to ACh in SHRSP as compared to WKY, suggesting a primary role for NO-dependent relaxation in SHRSP. To determine the isoform .responsible for the NO-dependent relaxation, · mesenteric arteries from SHRSP were incubated with selective iNOS and nNOS inhibitors. Vessels from SHRS.P exhibited a decreased relaxation response with iNOS inhibition. Western blot analysis revealed an increase in total eNOS, phospho-eNOS and iNOS levels in vessels from SHRSP. In addition, increases in phospho NF-KB were observed, suggesting a mechanism for increased iNOS protein expression. Overall, these data suggest a compen.satory role for NO by increased eNOS activation and iNOS expression. INTRODUCTION

Although hypertension is a multifactorial disease with varied etiologies, it

is widely accepted that increased peripheral vascular resistance and vascular wall inflammation play a primary role in the pathogenesis and progression. The

small diameter arteries and arterioles give rise to the majority of resistance in the

circulation and hence are the primary mediators of systemic blood pressure.

These vessels have been shown to be structurally and functionally altered in

hypertension. (1)

The vascular endothelium is crucial in the maintenance of proper vessel tone, releasing various vasoactive factors, including nitric oxide (NO),

prostacyclin and various endothelial hyperpolarizing factors (EDHF) and

contracting factors (EDCF). (2) Our understanding of the mechanisms by which the endothelium regulates these vasoactive factors is important in further

characterizing and treating this disease.

Nitric ·oxide is an important cell signaling molecule, which is kr1own to

mediate vascular smooth muscle cell relaxation. (3) While NO serves various

beneficial roles in the cell, as a second messenger and as a defense mechanism

during host invasion, excessive or deficient NO.production can lead to

deleterious results. Furthermore, decreased NO has been implicated in the

pathophysiology of hypertension, where decreased NO bioavailability leads to

increased vascular tone. In addition, prolonged, massive quantities of NO

produced during a pro-inflammatory state is known to exert various noxious

effects. Taken with the knowledge that this gas is easily diffusible and highly reactive, the biosynthesis of NO via nitric oxide synthase (4) is tightly regulated

and controlled. (5-6)

There are thr~e NOS genes, each located on different chromosomes

· leading to three distinct protein products; neuronal NOS (nNOS, also termed as

NOS1 ), inducible NOS (iNOS, also termed NOS2), and endothelial NOS (eNOS,

also termed NOS3). Under physiological conditions, NO is produced through the

5-electron oxidation L-arginine to L-citrulline, requiring various substrates and

cofactors. In their absence, NOS uncoupling occurs leading to decreased NO

production and increased reactive oxygen species (ROS) generation. (6)

Each of these isoforms pe.rforms a specific function in the cell.

Endothelial NOS is a constitutive, cytosolic enzyme, dnly producing small

amounts of NO in response to receptor or physical stimulation. However; iNOS is

activated in macrophages, endothelial or other cell types following feed-forward

cytokine stimulation, such as interleukin-1 (3, interferon-y and tumor necrosis

2 factors. (7) This isoform is not affected by changes in Ca + concentrations, nor

dependent upon the same cofa,ctors as eNOS. Additionally, iNOS is c~pable of

producing more than 1000-fold more NO than produced by eNOS for a sustained

period of time. The massive quantity of NO produced through this process is

potentially cytotoxic, although the mechanisms of this process are not well

known. (8-9)

In the_ central nervous system, NO is a neurotransmitter. The isoform,

· nNOS, is activated by glutamate acting upon the N-methyl-D-aspartate (NMDA)

2 receptor, causing a subsequent increase in Ca + concentration thus activating nNOS in the same Ca2+/calmodulin-dependent manner described previously.

Although constitutive as well, nNOS has been shown to be induced after

· stimulation or pathological insult. In the periphery, the non-adrenergic, non­

cholinergic (NANC) system performs various physiological roles in a NO­

dependent mann_er, including neurogenic vasodilation of the penile and airway

smooth muscle tissues.

The receptor for NO is soluble guanylate cyclase (sGC), and activation of the enzyme le~ds to increased cGMP production. (3) Cyclic GMP levels

increase to activate cGMP-dependent protein kinase I, which in turn

phosphorylates a number of protein targets including those which regulate vascular smooth muscle (VSM) tone ..

Given the physiological role of the three NOS isoforms and the importance

of NO in controlling vascular tone, we aimed to determine the prospective roles

of these enzymes in the· mesenteric arteries of the spontaneously hypertensive stroke prone rat (SHRSP) as compared to normotensive Wistar-Kyoto (WKY)

rats. We hypothesize that the SHRSP resistance vessels undergo a 'switch' and become more NO-dependent as compared to the WKY vessels. We also hypothesize that the increased pro-inflammatory milieu present in the

hypertensive rats leads to increased iNOS NO production, as compared to the

WKY. MATERIALS AND METHODS

Five month-old male stroke-prone spontaneously hypertensive rats

(SHRSP) were obtained from the breeding colony at the Michigan State

University. Age-matched male Wistar-Kyoto (WKY) rats were purchased from

Harlan (Indianapolis IN). Rats were maintained on a 12-hour light dark cycle, housed two per cage and allowed access to normal chow and water ad libitum.

Systolic blood pressure (SBP) was measured in non-anesthetized animals .by tail cuff using a RTBP1001 blood pressure system (Kent Scientific Corporation,

Connecticut, MA, USA).

All procedures were performed in accordance with the Guiding Principles in the Care and Use of Animals, approved by the Medical College of Georgia

Committee on the Use of Animals in Research and Education.

Functional Studies

After euthanasia with CO2, the mesentery was rapidly excised and bathed in ice-cold physiological salt solution (NaCl 120mM, KCI 4.7mM, KH2PO4

1.18mM, NaHCO3 14.9mM, dextrose 5.6mM, CaC'2·H2O 1.6"!1M). Second order mesenteric vessels were carefully isolated and mounted as ring preparations on two stainless steel wires (2:200 mm in length with internal diameter =100 to 150

µm) in a small-vessel myograph (Danish MyoTech, Aarhus, Denmark). Vessels were maintained at 37° and continuously aerated with 95% 02, 5% CO2 and allowed to stabilize for at least 45mins, at ·an optimal passive tension of 3.5mN.

After stabilization, tissues were contracted with KCI (120mM) solution to establish a reference for contractile responsiveness. To determine the viability of the

endothelium, contraction was stimulated via phenylephrine (Phe; 10µM) followed

by acetylcholine (ACh; 10µM). Vessels were then washed before performing

concentration-response curves (CRC) and after each CRC. To determine which

vasodilator pathway primarily functions in the mesenteric artery, CRC to ACh in

Phe contracted vessels was performed in the presence of vehicle or the following

inhibitors:. Nw-nitro-L-arginine methyl ester [L-NAME, NOS inhibitor; 100µM] or I . indomethacin [cyclooxygenase (COX) inhibitor; 10µM]. The remainder of

experiments w~s performed to determine which NOS isoform predominately

functions as the NO producer, and whether substrate availability is affecting NO·

production. The inhibitors- used to determine this were: 1400W (iNOS inhibitor;

100µM), 7-nitroindazole (nNOS inhibitor; 100µM), L-arginine {NOS

substrate; 10µM) and nor-NO HA (arginase inhibitor; 10µM) Force measurements

were collected using Chart™ Software (ADI Instruments, Colorado Springs, CO)

for Powerlab data acquisition .systems (ADI I.nstruments).

Tissue homogenization

·Following dissection and fat removal, mesenteric arteries from SHRSP

and WKY rats were quick frozen with liquid nitrogen, pulverized in a liquid

nitrogen-cooled mortar and pestle and solubilized in a 255 mM sucrose/1 0mM

Tris buffer (pH 7.4) with protease (0.5 mM phenylmethanesulphonylfluoride,

[PMSF], 2mM ethylene glycol tetraacetic acid [EGTA], 10 µg/ul aprotinin and 10

µg/ml leupeptin) and tyrosine phosphatase (1 mM sodium orthovanadate) inhibitors. Homogenates were centrifuged (3000g for 1 hour, 4°C) and supernatant total protein (Bio-Rad) measured.

Western Blotting

Total supernatant protein (40µg total, 4:1 in denaturing sample buffer and boiled for 5 minutes), was separated on a 10% SD_S-polyacrylamide gels (SDS­

PAGE) and transferred to lmmobilon-P nitrocellulose membrane. Membranes were blocked to rule out non-specific binding using 5% skim milk in Tris-buffered saline solution with Tween (0.1 %) for 1 hour at 24°C. Membranes were then probed _overnight (4°C) · with primary antibodies as follows: eNOS (1 :500), phospho-eNOS (1:1000), iNOS (1 :500),) Akt (1:1000), phospho~Akt (1:1000),

NF-KB (1: 1000), phospho-NF-KB (1 :500). All antibodies where purchased from

Cell Signaling. Blots were washed three times with tris-buffered saline plus

Tween- 20 (TBS-T) (30 minutes, 5 minutes, 5 minutes) an9 once with TBS (5 minutes). Depending · upon the antibody, either anti-rabbit or anti-mouse . horseradish peroxidase-linked secondary antibody (Amersham Labs) was added for one hour and incupated with the blots at 4°C. Blots were washed again using the described regimen. Enhanced chemiluminescence· (Super Signals Ultra,

Pierce) was used to visualize labeled bands. Densitometry was evaluated using

NIH Image J program.

Statistical Analysis For each agonist concentration-response curves were fitted using a nonlinear interactive fitting program (GraphPad Prism, Graph Pad Software Inc.,

San Diego CA), and values expressed as percent of maximal relaxation. Agonist potencies and maximum response are expressed as negative logarithm of the molar concentration of agonist producing 50% of the maximum response (pO2) and maximum effect elicited by the agonist (Emax), respectively. Data are expressed as means±SEM (n), where n is the number of experiments performed.

·statistical analysis of the concentration-response curves was performed by using two-way analysis of variance (ANOVA) for comparison between the groups.

Western blot data were analyzed by one-sample t test. Values of p<0.05 were considered a statistically significant differe_nce.

Drugs

L-Phenylephrine hydrochloride, acetylcholine, L-NAME, L-arginine an~. indomethacin;. 1400W and 7-nitroindazole were all purchased from Sigma­

Aldrich. nor-NOHA was purchased from Cayman Chemical. lndomethacin was dissolved in ethanol, and nor-NOHA was dissolved in DMSO. All other stock solutions were prepared by· using water. Control solutions containing vehicle levels of ethanol and DMSO were also used through the experimental protocol. RESULTS

Systolic blood pressure.

At 22 weeks, SHRSP displayed higher systolic blood pressure (mmHg), in comparison with WKY rats (208±6.3 vs. 121±2.7, n=12, respectively).

Mesenteric arteries from SHRSP exhibit increased dependence upon NO for endothelium-mediated relaxation.

Endothelium-dependent relaxation· responses were determined by CRC to

ACh, in small-mesenteric arteries. Surprisingly, there was no difference in sensitivity to ACh in vessels from SHRSP as compared to WKY controls (n=4,

Figure 1 a), as. evidenced by the pD2 values (8.03±2.3 vs 8.22±0.25 WKY, n=4 -

Table 1). A small decrease in maximal relaxation in mesenteric arteries from

SHRSP was observed (Emax 86.5±3.75% vs 101.3±2.3 %\J\/KY, n=4).

To determine the contribution of NOS and COX pathways in endothelium­ dependent relaxation, CRCs to ACh were performed in the presen_ce of L-NAME

(1 00µM) and indomethacin (1 0µM). As shown in Figure 1b, NOS inhibition with

L-NAME did not significantly reduce the relaxation response in WKY vessels; however, a significant rightward shift in the dose-response curve was observed in

L-NAME-treated SHRSP vessels as compared to L-NAME-treated WKY vessels

(Emax 62.5±11.5% vs. 100.7±6.2%, n=4, respectively). This effect was potentiated when SHRSP. vessels were simultaneously incubated_ with L-NAME and indomethacin (n=4, Figure 1 c). Similar to NOS only inhibition, the addition of indomethacin did .not change the ACh-mediated relaxation in WKY vessels (Emax 103.7±3.5%, n=4), but ACh-induced relaxation response was drastically decreased in arteries from SHRSP (Emax 39.1 ±1.1 %, n=4), suggesting an impaired production of EDHF in arteries from hypertensive rats. Together, these data suggest a role for NO and COX-derived metabolites in endothelium­ dependent relaxation in the mesenteric arteries from the SHRSP.

Mesenteric arteries from SHRSP depend upon iNOS for NO-mediated relaxation.

To investigate the NOS isoform which is responsible for the relaxation responses in the SHRSP vessels, CRCs were performed in the presence of vehicle or 7-nitroindazole, a selective nNOS inhibitor and 1400W, a selective iNOS inhibitor. As shown in Figure 2a, the relaxation response to ACh was not attenuated in WKY vessels with the nNOS.inhibitor as compared to WKY control vessels (Emax 103.8±3% vs. 103.2±2%, n=4, respectively). In SHRSP vessels, a small, but insignificant decrease in maximal relaxation was observed in those incubated with the nNOS inhibitor as compared to SHRSP controls (Emax

73.4±3.8%, vs. 92.7±1.7%, n=4, respectively) (Figure 2b). These data suggest that nNOS is not significantly contributing to the NO production in SHRSP.

CRCs were performed, in the presence of a selective iNOS inhibitor or vehicle.

The vessels from WKY normotensive rats exhibited no decrease in relaxation with iNOS inhibition (Emax 98.4±3.2%, vs. 92.1 ±2.5%, n=4, respectively - Figure

3a); however, iNOS inhibition in SHRSP vessels led to a significantly diminished capacity for vasorelaxation (Emax 95.8±1.2%, vs. 59.4±3.5%,n=4, respectively; p<0.05 - Figure 3b). These data suggest that iNOS is contributing, at least partially, to the ACh-mediated relaxation seen in SHRSP mesenteric arteries. Increased protein expression of iNOS, as well a·s increased NF-KB activation,

was observed in mesenteric arteries from SHRSP.

· To further investigate the mechanism .of NO-mediated relaxation in the

SHRSP.vessels, molecular experiments were performed. Total protein was used

in Western blot analysis from mesenteric arteries of WKY and SHRSP. In blots,

equally loaded with protein from both WKY and SHRSP mesenteric arteries, a

sign'ificant increase in iNOS protein expression was detected as compared to

WKY (n=4, p<0.05) as revealed in Figure 4, confirming the functional data ..

Knowing that nuclear fa·ctor kappa-light-chain-enhancer of activated B cells or

NF-KB and cytokines can lead to increased transcription and activation of iNOS,

membranes were probed for total· and phosphorylated NF-KB. (10-11) Although

total protein levels for NF-KB were not changed, an increase in phospho-NF-KB was observed (Figure 5).

Mesenteric arteries from SHRSP exhibit increased activation of eNOS and

upstream activator Akt.

Increased levels of total eNOS were detected in SHRSP mesenteric

arteries as compared to WKY. This was accompanied by an increased activation

state of eNOS as well (n=4, p<0.05 - Figure 6). Endothelial NOS activation was

determined by measuring the phosphorylation of eNOS at ser1177, a known site

for post-translational modification and subsequent activation of eNOS. The

upstream signaling proteins Akt, once activated, are.known to phosphorylate and·

activate eNOS. Mesenteric arteries from SHRSP did not express higher levels of Akt; however, an increased activation of Akt was detected as depicted by

phospho-Akt (WKY n=5, SHRSP n=4 - Figure 7).

Increasing substrate for eNOS by arginase inhibition or incubation with L­

arginine did not increase ACh-mediated relaxation in SHRSP mesenteric

vessels.

NOS utilizes the substrate, L-arginine to increase NO production. Given that these data suggest an increased NO production ·or sensitivity in SHRSP,· and

substrate availability for NO production was determined. CRCs were performed

in the presence of L-arginine and an arginase inhibitor (nor-NOHA). As shown in

Figure 8, no differences in relaxation responses were found in SHRSP (b) or

WKY (b) vessels incubated with L-arginine as compared to control (n=4). CRCs were next performed in the presence of the arginase inhibitor, nor-NOHA, to be

certain that increased arginase 8:Ctivity is not altering NOS substrate and thus NO

bioavailability. In vessels from SHRSP and WKY, no differences were observed

in vessels treated with the arginase inhibitor as compared to control (n=4 - Figure

9). Interestingly, the oscillatory activity of the SHRSP mesenteric arteries after

ACh-induced relaxation was blocked in vessels incubated with L-arginine (Figure

10). DISCUSSION

It is well known that resistance vessels mediate relaxation primarily in a

NO-independent manner, mostly via prostacyclin/COX-derived metabolites and

EDHF. EDHF dependence increases as the size of the vessel decreases. (12-

16) However, our data suggest a major role for NO in relaxation responses in

SHRSP resistance vessels,· which is most likely a compensatory mechanism accounting for lack of other vasodilators, i.e.-, EDHF_ and increase in EDCF. (17-

18)

Hypertension is a multi-factorial disease, and increased vascular tone is a component of the pathology. Endothelial dysfunction is well characterized in hypertension, and although controversial, there is asuggested alteration in NO production and/or metabolism. In genetic hypertension, an apparent endothelial dysfunction and impaired vasodilatory response to increased ·flow and receptor­ dependent agonists have been reported. (8, 19-21) The same has been found in th_e SHRSP, with a correlating decrease in ACh-mediated vasorelaxation,

· signifying that there may be decreased NO production, m~tabolism or bioavailability; a likely cause for the increased·tone. (7) However, there is still conflicting evidence within the literature, as some investigators have found decreased expression of NOS and a controversy still surrounds the mechanisms.

(7, 22-24) It is widely accepted that the primary vasodilators in small resistance arteries are EDHF and the prostaglandins, with NO dependence last.

Interestingly, normotensive, healthy vasculature is capable of compensating for lack of various vasodilators. This has been shown by using pharmacological inhibitors for NOS and COX, the use of NOS knock~out mice and evidence that various EDHFs activate an array of K+ channels in the VSM leading to hyperpolarization and relaxation. (3,-25) Our results in WKY support this View, as incubation with L-NAME and indomethacin did not attenuate the relaxation response. However, in the SHRSP vessels, the significant decrease in relaxation that occurred with NOS and COX inhibition suggests that mesenteric arteries from SHRSP rely upon different mechanisms for relaxation as compared to WKY.

Although there is conflicting evidence, other investigators have found similar results; increased NO production in in vivo and ex vivo preparations (26-28) and decreased EDHF production. (29) The primary mode of relaxation seems to be attributed to NO, signifying a switch in vasoactive mediators responsible for relaxation in SHRSP.

Given the increased production of NO in .SHRSP, as suggested by our data, we sought to investigate which of the NOS isoforms is primarily responsible for the relaxation response in SHRSP. We have shown that iNOS contributes to NO production in SHRSP mesenteric arteries. A probable mediator of the increased iNOS expression is NF-KB.(10) Corroborating these data are the increased levels of phos·pho-NF-KB detected locally in the mesenteric vessels of SHRSP. I We also report an increase in total eNOS expression and activation, in

addition to the increased phosphorylation and activation of the upstream eNOS

mediator Akt. In this study, we observed an increase in mesenteric eNOS

.expression and activation, suggesting that eNOS is playing a compensatory role

in the SHRSP resistance vessels. In addition, we observed an increased level of

phospho-Akt. Akt, when phosphorylated, becomes activated and eNOS is

accepted as a downstream effector.(30)

NO production from eNOS is moderate and short-lived, inducing changes·

in VSM tone. In this scenario, NO is beneficial, serving to enhance the VSM

relaxation and decrease platelet aggregation. In qontrast, the NO production by

iNOS is massive and prolonged, possible to the extent which could induce

cellular damage. Several investigators have shown a role for iNOS in

hypertension. Yen and colleagues demonstrated in SHR, that inhibition of iNOS

in vivo can suppress the development of hypertension, and demonstrated an

increase in iNOS expression the aorta. (5, 8, 32-34) A prominent theory is that

this sustained NO generation is contributing to the oxidant-mediated endothelium

dysfunction and that iNOS is a primary contributor to the ROS and NOS levels.·

(5, 35-36) Schiffrin, et al., and others have recently characterized hypertension to

be associated with elevated levels of circulating pro-inflammatory cytokines,

which are known activators of iNOS. (1, 24, 35, 37) This, coupled with the

increased ROS and superoxide anion that is present, a vicious circle is

. completed via the activation of NF-kB, thus further enhancing iNOS expression.

Other studies, showing enhanced iNOS expression, corroborate this hypothesis. (33, 38) Our data, showing that in ex vivo preparations from SHRSP, iNOS

inhibition significantly decreases ACh-mediated relaxation, also confirms the role

that iNOS plays in hypertension, as the WKY vessels did not exhibit a decreased

relaxation response. We further established this possibility by showing an

increased iNOS expression in the mesenteric arteries.

The increased iNOS expression seems to be increasing NO bioavailability

. and aiding eNOS in compensating for vascular dysfunction. It is accepted that

excessive NO production, especially in association with increased inflammatory

mediators where ROS are pres·ent, lead to peroxynitrite and superoxide

production. In that, we believe that the increased iNOS expression and NO

production is damaging to the endothelium and vasculature and is serving to

further increase cellular toxicity and endothelium injury in the SHRSP.

Additionally, we aimed to investigate if the substrate for NOS, L-arginine,

could play a role in the increased production of NO. In our experiments using L­

arginine and'the arginase inhibitor, nor-NOHA, we did not observe a difference in

the relaxation response in SHRSP mesenteric arteries. Although this was

surprising, as others have shown th~t an increase in NOS substrate can lead to

increases in receptor-mediated relaxation, it does not seem to be the case here.

Furthermore, these, data are difficult to interpret, since no ·functional data from

SHRSP has yet been pµblished regarding L-arginine metabolism. One

possibility seems to be that the mechanisms of gen·etic hypertension in the

SHRSP are different than that of other forms of hypertension, and that in the SHRSP mesenteric arteries, and in this case, substrate availability is not rate limiting.

Although L-arginine was not able to promote greater relaxation in mesenteric arteries from SHRSP, we observed that it was able to prevent oscillitary activity, observed after the arteries were contracted with Phe. (39)

This oscillatory ac~ivity is a phenomenon known to occur in resistan·ce arteries; however, the mechanisms are widely unknown. It is speculated that this event can occur.in an endothelium-dependent and independent mechanism, and relies on many factors. It is, however, generally accepted to be modulated, in some vascular beds by NO. (39-40) Vasomotion, which is endothelium­ dependent, relies upon NO for hyperpolarization of the vasculature, ·and closure

2 of voltage gated Ca + channels (VGCC) either directly through the activation of

2 large conductance Ca + activated K+ channels (BKca) or indirectly through protein kinase G (PKG). (39, 41-44)

· In conclusion, we have shown an increased dependence upon NO for relaxation in the mesenteric arteries of SHRSP as compared to their · normotensive counte.rparts, WKY. We observed increased levels of iNOS protein expression, as well as increased eNOS expression and activation levels, suggesting these isoforms as the mediators of the increased NO production.

ACKNOWLEDGEMENTS This study was supported by grants from FAPESP - Fundacao de Amparo a Pesquisa de Sao Paulo to RCT and the National ·Heart, Lung and Blood Institute at the National Institutes of Health [5R01 HL071138-08] to RCW.

REFERENCES:

1. Savoia C; Schiffrin EL. Inflammation in hypertension. Curr Opin Nephrol Hypertens. 2006 Mar;l5(2):152-8. 2. Vanhoutte PM, Shimokawa H, Tang EH, Feletou M. Endothelial dysfunction and vascular disease. Acta Physiol (Oxf). 2009 Jun; 196(2): 193-222. · 3. Vanhoutte PM. How we learned to say NO. Arteriosclerosis, Thrombosis and Vascular Biology. 2009;29. 4. Gil'yano NY, Stepanov SI, Konevega LV, Noskin LA. Intracellular NO concentration and its changes in carcinoma cells and cultured human endotheliocytes under the influence of inhibitors and inductors of nitric oxide synth~ses. Bull Exp Biol Med. 2007 Feb;143(2):248-50. 5. Pacheco ME, Marin J, Manso AM, Rodriguez-Martinez MA, Briones A, Salaices M, Redondo J. Nitric oxide synthase induction by ouabain in vascular smooth muscle cells from normotensive and hypertensive rats. J Hypertens. 2000 Jul;18(7):877-84. 6. Bryan NS BK, Murad F. Discovery of the nitric oxide signaling pathway and targets for drug development. Frontiers in Bioscience. 2009.; 14: 1-18. 7. Hirafuji M, Tsunoda M, Machida T, Hamaue N, Endo T, Miyamoto A, Minami M. Reduced expressions of inducible nitric oxide synthase and cyclooxygenase-2 in vascular smooth muscle cells of stroke-prone spontaneously hypertensive rats. Life Sci.· 2002 Jan 11 ;70(8):917-26 .. 8. Hong HJ, Loh SH; Yen MH. Suppression of the development of hypertension by the inhibitor of inducible nitric oxide synthase. Br J Pharmacol. 2000 Oct; 131 (3) :631-7. 9. Wu CC, Yen MH. Nitric Oxide Synthase in Spontaneously Hypertensive Rats. J Biomed Sci. 1997;4(5):249-55. 10. Kone BC, Schwebel J, Turner P, Mohaupt MG, Cangro CB. Role of NF-kappa B in the regulation of inducible nitric oxide synth8:se in an MTAL cell line. Am J Physiol. 1995 Nov;269(5 Pt 2):F718-29. 11. Mohaupt MG, Schwebel J, Elzie JL, Kannan GS, Kone BC. Cytokines activate inducible nitric oxide synthase gene transcription in inner medullary collecting duct cells. Am J Physiol. 1995 Apr;268(4 Pt 2):F770-7. 12. Komori K, Lorenz RR, Vanhoutte PM. Nitric oxide, ACh, and electrical and mechanical properties of canine arterial smooth muscle. Am J Physiol. 1988 Jul;255 (1 Pt 2):H207-12. 13-. Urakami-Harasawa L, Shimokawa H, Nakashima M, Egashira K, Takeshita A. Importance of endothelium-derived hyperpolarizing factor in human arteries. J Clin Invest. 1997 Dec 1;100(11):2793-9. 14. Feletou M, Vanhoutte PM. Endothelium-dependent hyperpolarization of canine coronary smooth muscle. Br J PharmacoL 1988 Mar;93(3):515-24. 15. Luksha L, Agewall S, Kublickiene K. Endothelium-derived hyperpolarizing factor in vascular physiology and cardiovascular disease. Atherosclerosis. 2009 Feb;202(2):330-44. 16. Tomioka H, Hattori Y, Fukao M, Sato A, Liu M, Sakuma I, Kitabatake A, Kanno M. Relaxation in different-sized rat blood vessels mediated by endothelium-derived hyperpolarizing factor: importance of processes mediating precontractions. J Vase Res. 1999 Jul-Aug;36( 4):311-20. · 17. Vanhoutte PM TE. Endothelium-dependent contractions: when a good guy turns bad! J Physiol. 20ff8;586:5295-304. 18. _ F eletou M, Vanhoutte PM. Endothelial dysfunction: a multifaceted disorder (The Wiggers Award Lecture). Am J Physiol Heart Circ Physiol. 2006 Sep;291(3):H985-1002. 19. Boulanger CM, Heymes C, Benessiano J, Geske RS, Levy BI, Vanhoutte PM. Neuronal nitric oxide synthase is expressed in rat vascular smooth muscle cells: activation by angiotensin 11 in hypertension. Circ Res. 1998 Dec 14-28;83(12):1271-8. 20. Kawata M, Mizutani T, Shimizu M, Takenaka K, Kobayashi S, Okada T, Miyamoto Y, Kanazawa K, Akita H, Yokoyama M. Endothelial nitric oxide synthase gene mutation and human leukocyte antigen analyzed in three cases of familial. vasospastic angina pectoris. Jpn Circ J. 2000 Jul;64(7):524-7. 21. Ebrahimian T, Mathieu E, Silvestre JS, Boulanger CM. Intraluminal pressure . incre3:ses vascular neuronal nitric oxide synthase expression. J Hypertens. 2003 May;21(5):937-42. 22. Malinski T, Kapturczak M, Dayharsh J, Bohr D. Nitric oxide synthase activity in genetic hypertension·. Biochem Biophys Res Commun. 1993 Jul 30;194(2):654-8. 23. Dubois G. Decreased L-arginine-nitric oxide pathway in cultured myoblasts from spontaneously hypertensive versus normotensive Wistar-Kyoto rats. FEBS Lett. 1996 Sep 2;392(3):242-4. 24. Singh A, Sventek P, Lariviere R, Thibault G, Schiffrin EL. Inducible nitric oxide synthase in vascular smooth muscle cells from prehypertensive spontaneously · hypertensive rats. Am J Hypertens. 1996 Sep;9(9):867-77. ·2s. Feletou M, Vanhoutte PM. EDHF: an update. Clin Sci (Land). 2009 Aug;l 17(4):139-55. 26. Junquero DC, Schini VB, Scott-Burden T, Vanhoutte PM. Enhanced production of nitric oxide in aortae from spontaneously hypertensive rats by interleukin-I beta. Am J Hypertens. 1993 Jul;6(7 Pt 1):602-10. 27. Xiao J, Pang PK. Does a general alteration in nitric oxide synthesis system occur in spontaneously hypertensive rats? Am J Physiol. 1994 Jan;266(1 Pt 2):H272-8. 28. Vaziri ND, Ni Z, Oveisi F. Upregulation of renal and vascular nitric oxide synthase in young spontaneously hypertensive rats. Hypertension. 1998 Juri;31(6):1248- 54. 29. Giachini FR, Carneiro FS, Lima VV, Carneiro ZN, Dorrance A, Webb RC, Tastes RC. Upregulation of intermediate calcium-activated potassium channels counterbalance the impaired endothelium-dependent vasodilation in stroke-prone spontaneously hypertensive rats. Transl Res. 2009 Oct; 154( 4): 183-93. 30. Sessa WC. eNOS at a glance. J Cell Sci. 2004 May 15;1 l 7(Pt 12):2427-9. · 31. Chou TC, Li CY, Wu CC, Yen MH, Ding YA. The inhibition by dantrolene ofL- arginine transport and nitric oxide _synthase in rat alveolar macrophages. Anesth Analg.· 1998 May;86(5):1065-9. 32. Aras-Lopez R,_Blanco-Rivero J, Hernanz R, Briones AM, Rossoni LV, Ferrer M, Salaices M, Balfagon G. Chronic ouabain treatment increases the contribution of nitric oxide to endothelium-dependent relaxation. J Physiol Biochem. 2008 Jun;64(2):115-25. 33. Chou TC, Yen MH, Li CY, Ding YA. Alterations of nitric oxide synthase expression with aging and hypertension in rats. Hypertension. 1998 Feb;31 (2):643-8. 34. Rossoni LV, Salaices M, Miguel M, Briones AM, Barker LA, Vassallo DV, Alonso MJ. Ouabain-induced hypertension is accompanied by increases in endothelial vasodilator factors. Am J Physiol Heart Circ Physiol. 2002 Nov;283(5):H2110-8. 35. Marin J, Rodriguez-Martinez MA. Role of vascular nitric oxide in physiological and pathological conditions. Pharmacol Ther. 1997 Aµg;75(2): 111-34. 36. Ginnan R, Guikema BJ, Halligan KE, Singer HA, Jourd'heuil D. Regulation of smooth muscle by inducible nitric oxide synthase and NADPH oxidase in vascular proliferative diseases. Free Radie Biol Med. 2008 Apr 1;44(7):1232-45. 37. Vaziri ND, Rodriguez-Iturbe B. Mechanisms of disease: oxidative stress and inflammation in the pathogenesis of hypertension. Nat Clin Pract Nephrol. 2006 Oct;2(10) :5 82-93. 38. Wu CC, Hong HJ, Chou TC, Ding YA, Yen MH. Evidence for inducible nitric oxide synthase in spontaneously hypertensive rats. Biochem Biophys Res Commun. 1996 Nov 12;228(2):459-66. 39. Yuill KH, McNeish AJ, Kansui Y, Garland CJ, Dora KA. Nitric Oxide Suppresses Cerebral Vasomotion by sGC-Independent Effects on Ryanodine Receptors and Voltage-Gated Calcium Channels. J Vase Res. 2009 Sep 4;47(2):93-107. 40. Rahman A, Hughes A, Matchkov V, Nilsson H, Aalkjaer C. Antiphase oscillations of endothelium and smooth muscle [Ca2+]i in vasomotion ofrat mesenteric small arteries. Cell Calcium. 2007 Dec;42(6):536-47. 41. Archer SL, Huang JM, Hainpl V, Nelson DP, Shultz PJ, Weir EK. Nitric oxide and cGMP cause vasorelaxation by activation of a charybdotoxin-sensitive K channel by cGMP-dependent protein kinase. Proc Natl Acad Sci USA. 1994 Aug 2;91(16):7583-7. 42. Mistry DK, Garland CJ. Nitric oxide (NO)-induced activation of large conductance Ca2+-dependent K + channels (BK(Ca)) in smooth muscle cells isolated from the rat mesenteric artery. Br J Pharmacol. 1998 Jul;124(6):l 131-40. 43. Bolotina VM, Najibi S, Palacino JJ, Pagano PJ, Cohen RA. Nitric oxide directly activates calcium-dependent potassium channels in vascular smooth muscle. Nature. 1994 Apr 28;368(6474):850-3. 44. Robertson BE, Schubert R, Hes_cheler J, Nelson MT. cGMP-dependent protein kinase activates Ca-activated K channels in cerebral artery smooth muscle cells. Am J Physiol. 1993 Jul;265(1 Pt l):C299-303. Figure 1. In SHRSP mesenteric arteri~s, nitric oxide is the primary vasodilator, as compared to WKY. Concentration response curves to ACh in second order mesenteric arteries from SHRSP as compared to WKY controls. Relaxation responses to ACh were assessed in vessels from SHRSP and WKY rats incubated with (a) vehicle, (b) L-NAME (NOS inhibitor, l0OµM) and (c) indomethacin (COX inhibitor, f0µM) plus L-NAME (1 0o'µM). Relaxation responses were calculated relative to maximal contraction produced from U-46619. Results are presented as mean±SEM of n=4 for each experimental group. *p<~.05. -

Figure 2. Neuronal NOS does not contribute to endothelium-mediated relaxation in SHRSP or WKY mesenteric arteries. Concentration response curves to ACh in second order mesenteric arteries from SHRSP as compared to WKY controls. Relaxation responses to ACh were assessed in vessels from (a) WKY or (b) SHRSP incubated with the nNOS inhibitor, 7-nitroindozole (lO0mM) or vehicle. Relaxation responseswere calculated relative·to maximal contraction produced from U-46619. Results are presented as mean±SEM ofri=4 for each experimental group. *p<0.05. .

Figure 3. Inducible NOS partially contributes to endothelium-mediated relaxation in SHRSP mesenteric arteries. Concentration response curves to ACh in s_econd order mesenteric arteries from SHRSP as compared to WKY controls. Relaxation responses to ACh were assessed in vessels from (a) WKY or (b) SHRSP incubated with the iNOS inhibitor, 1400W (lO0mM) or vehicle. Relaxation responses were calculated relative to . maximal contraction produced from· U-4119. Results are presented as mean±SEM of n=4 for each experimental group. *p<0.05.

Figure 4. Mesenteric arteries from. SHRSP exhibit increased iNOS expression. (a) Representative Western blots for iNOS (131kDa) and P-actin (42kl)a). Bar graph representing densitized values (in arbitrary units, AU) for protein expression of (b) iNOS in mesenteric arteries of WKY and SHRSP. All values were normalized to ~-actin and expressed as mean±SEM, n=4. *p<0.05

Figure 5. Mesenteric arteries from SHRSP exhibit increased expression of phosphorylated NF-KB. (a) Representative Western blots for NFKB (65kDa), phospho­ NFKB (65kDa) and P-actin ( 42kDa). Bar graph representing densitized values (in arbitrary units, AU) for protein expression of (b) phospho-NF1d3 and (c) NFKB in mesenteric arteries of WKY and SHRSP. All v~lues were normalized to ~-actin and expressed as mean±SEM, n=4. *p<0.05

Figure 6. Mesenteric arteries from SHRSP exhibit increased expression of phosphorylated eNOS. (a) Representative Western blots for eNOS (1_40kDa), phospho­ eNOS (140kDa) and ~-actin (42k.Da). Bar graph representing densitized values (in arbitrary units, AU) for protein expression of (b) eNOS and (c) phospho-eNOS in mesenteric arteries of WKY and SHRSP. All values were norinalized to ~-actin and expressed as mean±SEM, n=4. *p<0.05 Figure 7. Mesenteric. arteries from SHRSP exhibit increased expression of phosphorylated Akt. (a) Representative Western blots for Akt (50kDa), phospho-Akt (75kDa), and P-actin (42kDa). Bar graph representing densitized values (in arbitrary units, AU) for protein expression of (b) Akt and (c) phospho-Akt in mesenteric arteries of WKY and_SHRSP. All values were normalized to P-actin and expressed as mean±SEM, WKY n=S, SHRSP n=4. *p<0.05

Figure 8. Incubation with L-a·rginine does not increase relaxation in mesenteric arteries from either WKY or SHRSP. Concentration response curves to ACh in second order mesenteric arteries from SHRSP as compared to WKY controls. Relaxation responses to ACh were assessed in vessels from (a) WKY or (b) SHRSP incubated with the NOS substrate, L-arginine, (1 0µM) or vehicle. Relaxation responses were calculated relative to maximal contraction produced from U-4119. Results_are presented as · mean±SEM of n=4 for each experimental group.

Figure 9. Arginase inhibition does not increase relaxation response in mesenteric arteries from neither WKY nor SHRSP. Concentration response curves to A Ch in _second order mesenteric arteries from SHRSP as compared to WKY controls. Relaxation responses to ACh were assessed in vessels from (a) WKY or (b) SHRSP incubated with the arginase inhibitor, nor-NO HA (1 0µM) or vehicle. Relaxation responses were calculated relative to maximal contraction produced from U-4119. Results are presented as mean±SEM of n=4 for each experimental group.

Figure 10. Incubation with L-arginine blocks oscillatory activity in SHRSP mesenteric arteries after ACh-induced relaxation. Second order mesenteric arteries from SHRSP were contracted to phenylephrine (1 0µM) and relaxation to acetylcholine induced. SHR.SP vessels reacted with oscillatory activity, whereas vessels incubated with L-arginine did not exhibit the same oscillatory activity. Figures 1a, b and c.

(a).

0 □ WKY

C: 25 ■ SHRSP 0 :.:; 50 ~ cu c6 0:: 75 ·= ~ 100

125 I I I I I I -9 -8 -7 -6 -5 -4 ACh (Log[M])

(b).

0 □ WKY

C: 25 .■ SHRSP 0 :.:; cu 50 >

125 I I I I I I -9 -8 -7 -6 -5 ;.4 AC h ( Log [Ml)

L-NAME (c).

0 □ WKY

C 25 ■ SHRSP 0 ;; 50 ~ -a; 75 ~ ~ 100

125 I I I I I I -9 -8 -7 -6 -5 -4 ACh (Log[M]) . ·· L-NAME and lndomethacin Figures 2a and b.

(a).

0 □ WKY

C: 25 0 WKY + 7-nitro 0 .:; 50 ~ n:s cii 75 ~ ~ 0 100

125 I I _I I I I -9 -8 . -7 -6 -5 -4 AC h ( Log[M])

(b).

0 111 SHRSP c: 25 • SHRSP + 7-nitro 0 ;. ~ -50 n:s a; 75 ~ ~ 0 100

125 I I I I I I -9 -8 -7 -6 -5 -4 ACh (Log[M]) Figures 3a and b.

(a).

0 □ WKY 25 0 WKY + 1400W .oC: ;:; 50 ~ cu -a; 75 ~ ';ff. 100

125 I I I I I I -9 -8 -7 -6 -5 -4 ACh (Log[M])

(b).

0 ■ SHRSP

C: 25 • SHRSP + 1400W 0 ;:; cu 50 >

125 I I I I I I -9 -8 -7 -6 -5 -4 ACh (Log[M]) Figures 4a and b.

(a)

WKY SHRSP

131kDa i-NOS

42kDa p-actin

iNOS 1.5

C: ;i. * ~ 1.0 c:c..I -·a;C: 0... 0.5 D..

0.0------=WKY SHRSP Figures Sa, band c.

(a).

WKY SHRSP

65kDa NFKB

65kDa phospho-NFKB

42kDa J3-actin

(b).

NFkB 0.3

C ;; ~ 0.2 cc.I C· -"cii 0... 0.1 C.

0.0...... ______1 WKY · SHRSP (c).

phospho-NFkB 0.7 0.6 * C ;; 0.5 () cu ~0.4 C: ·a; 0.3 0 c. 0.2 0.1 0.0 WKY SHRSP Figures 6a, b and c.

(a). WKY

140kDa e-NOS

phospho-eNOS 140kDa

~-actin 42kDa

(b).

eNOS 2

C:

·~ ns c:c.I "2 1 ·a;

0I., D.. o------,--...1--...J WKY SHRSP (c).

e-NOS (SER1177) 1.0 *

C ;; (.) cu I cQ. c: 0.5 ·a;...... 0 0. o.o------WKY SHRSP Figures 7a, band c.

(a).

WKY SHRSP

50kDa Akt

75kDa phospho-Akt

42kDa

(b).

Akt 1.25

C 1.00 t; l 0.1s c ·;e 0.50 c. 0.25

0.00.._...... ______

1 WKY SHRSP (c).

Akt-phospho 1.00

c· ~ 0.75 (.) co.. c 0.50 ·;' 0 a: 0.25

WKY SHRSP Figures 8a and b.

(a).

0 □ WKY C: 25 0 WKY + L-arginine 0 ; ca 50 >

125 I I I I I I -9 -8 -7 -6 -5 -4 ACh (Log[M])

(b).

0 ■ SHRSP

C 25 • SHRSP + L-arginine 0 ; ca. 50 >

125 I I I I I i -9 -8 -7 -6 -5 -4 ACh (Log[M]) Figures 9a and b.

(a).

0 □ WKY

C 25 0 WKY + NOR-NOHA 0 ; 50 ~cu cii 0:: 75 ~0 100

125 I I I I I i -9 -8 -7 -6 ".'5 -4 ACh (Log[M])

(b).

0 ■ SHRSP

C 25 • SHRSP + NOR-NOHA 0 ; 50 ~cu cii 0:: 75 ~ 100

125 I I I I I i -9 -8 -7 -6 -5 -4 ACh (Log[M]) Figure 10.

Pe ACh

Pe ACh T

a I,...... , ....

L-Arginine ·. Table 1. Emax and pD2 values for ACh in resistance mesenteric arteries from SHRSP and WKY rats

WKY SHRSP

Emax pD2 Emax pD2

Ach 101.3±2.3% 8.03±0.07 86.55±3. 75% 8.23±0.25

ACh + L-NAME 100.7±6.2% 8.26±0.6 62.5±5.3% 10.22±5

ACh + L-NAME + 103±3.5% 7.84±0.15 39±1.0% 7.80±0.11 I ndomethacin

7-nitroindozole 103.2±2.0% 8.86±0.72 73.4±3.8% 8.38±0.5

1400W 92.1±2.5% 7.93±0.09 59.4±3.5% 7.68±0.16

Values are expressed as means±SEM for n= 12. Maximum effect elicited by the agonist (Emax) and negative logarithm of the molar concentration of agonist producing 50% of the maximum response (pD2). * indicates p<0.05 vs. vehicle.