USF Libraries
USF Digital Collections

Multi-level regulation of argininosuccinate synthase

MISSING IMAGE

Material Information

Title:
Multi-level regulation of argininosuccinate synthase significance for endothelial nitric oxide production
Physical Description:
Book
Language:
English
Creator:
Corbin, Karen Davidowitz
Publisher:
University of South Florida
Place of Publication:
Tampa, Fla
Publication Date:

Subjects

Subjects / Keywords:
Insulin
Vascular endothelial growth factor
Bradykinin
Ceramide
Kinase
Dissertations, Academic -- Molecular Medicine -- Doctoral -- USF   ( lcsh )
Genre:
non-fiction   ( marcgt )

Notes

Summary:
ABSTRACT: The citrulline-nitric oxide (NO) cycle, comprised of the enzymes argininosuccinate synthase (AS), argininosuccinate lyase (AL) and endothelial nitric oxide synthase (eNOS), is responsible for the regulated production of endothelial NO. Although most studies have focused on eNOS to uncover important regulatory mechanisms, we and others have determined that AS is an essential and regulated step in endothelial NO production. AS is rate limiting for endothelial NO production and is the primary source of arginine, the substrate for eNOS-mediated NO production, despite saturating intracellular levels of arginine and available arginine transport systems. AS is essential for endothelial cell viability and its expression is regulated coordinately with eNOS by TNF and thiazolidenediones with concomitant effects on NO production. Given the importance of AS for endothelial health, we explored three independent regulatory mechanisms.In Chapter One, the functional consequences of altered AS expression due to overexpression, insulin, VEGF and ceramide were studied. We demonstrated that overexpression of AS leads to enhanced NO production and that insulin, VEGF and ceramide coordinately regulate the expression of AS and eNOS. In Chapter Two, the first post-translational modifications of AS in the endothelium were characterized. We determined that AS is an endogenous phosphoprotein in the endothelium, described several levels of biological significance of AS phosphorylation, identified 7 sites of AS phosphorylation and began to uncover the direct impact of phosphorylation on AS function. Finally, in Chapter Three, endothelial AS subcellular localization was defined and important protein interactions were identified including caveolin-1 and HSP90.The work presented in this dissertation demonstrates that multiple mechanisms regulate the function of AS, often coordinately with eNOS, and have a direct impact on nitric oxide production. Our findings suggest that the global understanding of the citrulline-NO cycle as a metabolic unit will unravel new paradigms that will re-define our understanding of the regulation of vascular function by NO.
Thesis:
Dissertation (Ph.D.)--University of South Florida, 2008.
Bibliography:
Includes bibliographical references.
System Details:
Mode of access: World Wide Web.
System Details:
System requirements: World Wide Web browser and PDF reader.
Statement of Responsibility:
by Karen Davidowitz Corbin.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains 228 pages.
General Note:
Includes vita.

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 002004904
oclc - 352927639
usfldc doi - E14-SFE0002692
usfldc handle - e14.2692
System ID:
SFS0027009:00001


This item is only available as the following downloads:


Full Text
xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam Ka
controlfield tag 001 002004904
003 fts
005 20090527162241.0
006 m||||e|||d||||||||
007 cr mnu|||uuuuu
008 090527s2008 flu s 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0002692
035
(OCoLC)352927639
040
FHM
c FHM
049
FHMM
090
QD415 (Online)
1 100
Corbin, Karen Davidowitz.
0 245
Multi-level regulation of argininosuccinate synthase :
b significance for endothelial nitric oxide production
h [electronic resource] /
by Karen Davidowitz Corbin.
260
[Tampa, Fla] :
University of South Florida,
2008.
500
Title from PDF of title page.
Document formatted into pages; contains 228 pages.
Includes vita.
502
Dissertation (Ph.D.)--University of South Florida, 2008.
504
Includes bibliographical references.
516
Text (Electronic dissertation) in PDF format.
520
ABSTRACT: The citrulline-nitric oxide (NO) cycle, comprised of the enzymes argininosuccinate synthase (AS), argininosuccinate lyase (AL) and endothelial nitric oxide synthase (eNOS), is responsible for the regulated production of endothelial NO. Although most studies have focused on eNOS to uncover important regulatory mechanisms, we and others have determined that AS is an essential and regulated step in endothelial NO production. AS is rate limiting for endothelial NO production and is the primary source of arginine, the substrate for eNOS-mediated NO production, despite saturating intracellular levels of arginine and available arginine transport systems. AS is essential for endothelial cell viability and its expression is regulated coordinately with eNOS by TNF and thiazolidenediones with concomitant effects on NO production. Given the importance of AS for endothelial health, we explored three independent regulatory mechanisms.In Chapter One, the functional consequences of altered AS expression due to overexpression, insulin, VEGF and ceramide were studied. We demonstrated that overexpression of AS leads to enhanced NO production and that insulin, VEGF and ceramide coordinately regulate the expression of AS and eNOS. In Chapter Two, the first post-translational modifications of AS in the endothelium were characterized. We determined that AS is an endogenous phosphoprotein in the endothelium, described several levels of biological significance of AS phosphorylation, identified 7 sites of AS phosphorylation and began to uncover the direct impact of phosphorylation on AS function. Finally, in Chapter Three, endothelial AS subcellular localization was defined and important protein interactions were identified including caveolin-1 and HSP90.The work presented in this dissertation demonstrates that multiple mechanisms regulate the function of AS, often coordinately with eNOS, and have a direct impact on nitric oxide production. Our findings suggest that the global understanding of the citrulline-NO cycle as a metabolic unit will unravel new paradigms that will re-define our understanding of the regulation of vascular function by NO.
538
Mode of access: World Wide Web.
System requirements: World Wide Web browser and PDF reader.
590
Advisor: Duane C. Eichler, Ph.D.
653
Insulin
Vascular endothelial growth factor
Bradykinin
Ceramide
Kinase
690
Dissertations, Academic
z USF
x Molecular Medicine
Doctoral.
773
t USF Electronic Theses and Dissertations.
4 856
u http://digital.lib.usf.edu/?e14.2692



PAGE 1

Multi-Level Regulation Of Ar gininosuccinate Synthase: Significance For Endothelia l Nitric Oxide Production by Karen Davidowitz Corbin A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy Department of Molecular Medicine College of Medicine University of South Florida Major Professor: Duane C. Eichler, Ph.D. Denise R. Cooper, Ph.D. William R. Gower, Ph.D. Mark P. McLean, Ph.D. Gene C. Ness, Ph.D. Date of Approval: November 17, 2008 Keywords: insulin, vascular e ndothelial growth factor, br adykinin, ceramide, kinase, phosphorylation, post-translational modificat ions, heat shock pr otein 90, caveolin, endothelial nitric oxide s ynthase, subcellular localiza tion, protein interactions Copyright 2008 Karen Davidowitz Corbin

PAGE 2

DEDICATION First and foremost, I would like to dedicate this work to God. My prayer is that He can use the gifts He has given me as a scientist and educator to make positive contributions to our society. I would also like to dedicate this work to my family. My mom, Aurora, dad, Bob, and brother, Bobby, ar e my biggest fans and have supported me in all my endeavors. They have always inspir ed me to do my best and reach for the stars. My husband, best friend and chefStephenhas been a huge support during these years of sacrifice. Despite all th e craziness of graduate school, he still chose to marry me. He has earned this degree just as much as I have. All my familyfrom The Davidowitz side and The Rodriguez sidehave contributed to my life through their love, support and guidance. Finally, my newest familyThe Corbinshave been such a wonderful addition to my life and have been rooting for me and supporting me from the day I met them. I do want to especially dedicate this work to my brav e and precious nephew Ethan Corbin. He was diagnosed with neuroblastoma at six months of age and has fought like a brave little soldier to beat that disease. He is an inspiration and a survivor. If anyone ever wonders why someone would dedicate their life to science and h ealthcare, it is for little angels like him. Without the advances in medicine th at have been made possible by scientific research, the gift of his life and love might have ceased to exist.

PAGE 3

ACKNOWLEDGEMENTS The scientific journey through graduate school is one full of many experiences. Some are expected and some are unexpected. Through the blo od, sweat and tears, it takes an army of people to teach you the lessons that you never knew you would need to learn. Graduate school is a time where you see the significance of ethics and learn the most valuable skill in scienceindependent thi nking. I certainly could not have made it through without the investment that many people made in me. Dr. Eichler is a fantastic mentor. From the day I walked in his office unexpectedly and asked him if I could join his lab, he has treated me like a colleague and not a student. He always valued my opinions (even if he never actually asked for them). He allowed me the intellectual freedom to build my proj ect from the ground up. Nothing can compare to the experience of figuring thi ngs out for yourself. He never pushed me because I push myself more than anyone, but he gently made suggestions that I ev entually realized I should follow. One of the things I value about him the most is his at tention to ethics. He never jumps to conclusions and has a true desire to make contributions to science that are based on truth. His concern is with improving the health of people, not pushing his own research agenda.

PAGE 4

I am grateful to all the members of the Eichler Lab that became such a huge part of my graduate career. First, the Eichler a nd Solomonson labs have functioned as a single unit for many years. So, I consider Dr. Solom onson (Dr. Sol) a co-mentor. He was always available to help me with scientific questions He was a great person to go to when things were all jumbled and I needed a clean-cut so lution. He often came into the lab and said: So, what have you discovered this week? That level of confidence in me was extremely motivating. The Eichler lab l iterally took me in with ope n arms and did everything possible to make my transition into their lab a smooth one. I would like to thank Bonnie Goodwin, Brenda Flam and Laura Pendleton for their help and support when I first entered the lab. Their scientific contributions in our lab set an excellent foundation for the development of my projects. I especially want to thank Laur a Pendleton. We went through a lot of transitions togetherfrom de partment mergers to grant issues to moving into a new lab. Then, in between that, there was the science. Through it all, Laura was the one who kept me calm and helped me with both science and life issues. Finally, our newest lab member, Sandi Shriver, has been a huge, huge help in th ese last few months. Her willingness to help with all aspects of keeping our lab running has been more than I could have ever expected. She is an extrem ely bright person that works tirelessly. She will be successful in whatever path she chooses. I would also like to recognize the cont ributions of my committee members: Denise Cooper, PhD, William Gower, PhD, Mark McLean, PhD, and Gene Ness, PhD. They were a very supportive committee that I could always count on to make it to meetings and give me good suggestions. They have cheered me on, pushed me and

PAGE 5

guided me. I would also like to thank Dr. Chris Baylis from the University of Florida for so graciously agreeing to serve as the outside chair of my committee. Thank you to all the collaborators that ha ve contributed to this work: John Koomen, PhD, and Vicky Izumi at the Moffitt Proteomics Core, Marina Tran and Jasbinder Sanghera, PhD, at SignalChem and Wa yne Guida, PhD, and Daniel Santiago at the USF Department of Chemistry. I would like to thank a few colleagues and friends that have known me for many years. Cathy Levenson, PhD, and Jodee Dorse y, PhD, at Florida State University have become true life-long mentors. Anne Brezina and all the dietitians at the James A. Haley Veterans Hospital and the Tampa & Pinellas Dietetic Associations have been such wonderful friends and cheerleaders over the y ears. I am grateful to all my friends and colleagues at The Heart and Vasc ular Institute of Florida. Th e experiences I gained there will be an asset to my career forever and th at was the birth place of my love of the science behind heart disease, diabetes and obesity. Importantly, I definitely could not have made it without a little help from my friends. My best girlfriends, bridesmaids and sisters Paula Calabrese, Kelli Carr, Megan Orseck and Megan Sheiman have always been there for me. I am truly blessed for their impact in my life. I have many other wonderful friends that I am not mentioning individually but who are very important to me. Then there were many people who helped carry me through the days I thought I was not going to make it through graduate school.

PAGE 6

It all began with Yira Bermudez and Shawna Shirley Gilman. We met in Foundations in Biomedical Sciences. We were so young, c onfused and nave. Then we navigated through all the phases of graduate school toge ther and helped each other through the good and bad times. Later in graduate school Thomas Lendrihas joined our group. He definitely made graduate school bearable w ith his humor and unique point of view. For all the others who crossed my path thanks! Each one of you touched my life and made a difference. One of my greatest hopes is that I can do for others what all the people in my life have done for me.

PAGE 7

i TABLE OF CONTENTS List of Tables vi List of Figures vii List of Abbreviations ix Abstract xi Introduction 1 Nitric Oxide: Historical Pe rspectives and Biochemistry 1 The Citrulline-Nitric Oxide Cycle 2 Endothelial Nitric Oxid e and Vascular Health 6 Endothelial Dysfunction 8 Prevention of Vascular Disorders 12 Argininosuccinate Synthase Function s Related to Nitric Oxide Production 12 Regulation of Endothelial NO Production: Expression 14 Regulation of Endothelial NO Production: Post-Translational Modifications 21 Regulation of Endothelial NO Productio n: Subcellular Localization and Protein Interactions 26 References 32 Specific Aims 55 Purpose 55

PAGE 8

ii Central Question and Hypothesis 55 Specific Aim 1 56 Specific Aim 2 56 Specific Aim 3 57 Working Model 57 References 60 Chapter One: Argininosuccinate Synt hase Function and Expression 62 Overview 62 Materials and Methods 64 Bovine Endothelial Cell Culture 64 AS Expression Vector 64 AS Overexpression 67 Western Blot 68 RNA Isolation and Real Time PCR 68 Luciferase Vector Construction 69 Luciferase Assays 69 AS Promoter Analysis 70 Nitric Oxide Assays 70 Statistical Analyses 71 Results 71 AS Overexpression Enhances Endothelial NO Production 71 AS Expression and Function are Enhanced by Insulin 75

PAGE 9

iii VEGF Regulates AS Expression and Enhances Endothelial NO Production 82 Ceramide Diminishes AS and eNOS Expression and Suppresses eNOS Activation 84 Discussion 86 References 93 Chapter Two: Argininosuccina te Synthase Phosphorylation 98 Overview 98 Materials and Methods 99 Bioinformatics 99 Bovine Aortic Endothelial Cell Culture 100 In Vivo 32 P Orthophosphate Labeling 100 Immunoprecipitation and Western Blot 101 Affinity Chromatography 102 Purification of Bovine Ar gininosuccinate Synthase 103 In Vitro Kinase Screen 104 Nitric Oxide Assays 105 Generation of AS Variants and Transient Transfections 105 Purification of Overexpressed AS 107 Liquid Chromatography Tandem Mass Spectrometry 108 In Silico Modeling of AS Three Dimensional Structure 109 Statistical Analyses 109 Results 110

PAGE 10

iv AS is an Endogenous Phosphoprotein 110 Biological Relevance of AS Phosphorylation 116 Identification of Specific Sites of AS Phosphorylation 124 Mechanism of AS Regulation by Phosphorylation 136 Discussion 144 References 156 Chapter Three: Argininosuccinate Synthase Subcellular Localiza tion and Protein Interactions 165 Overview 165 Materials and Methods 166 Immunofluorescence 166 Immunoprecipitation and Protein Identification Using LC-MS/MS 166 Bioinformatics 167 Results 167 AS Subcellular Localization Overlaps with eNOS and Caveolin-1 167 AS Protein Interactions 170 Proteomic Examination of the Nitric Oxide Metabolome 173 Discussion 177 References 188 Perspectives 194 Summary 194 Significance 195 Limitations 199

PAGE 11

v Future Directions 200 References 203 Appendix A: Related Publications 205 Troglitazone up-regulates vascular e ndothelial argininosuccinate synthase About the Author End Page

PAGE 12

vi LIST OF TABLES Table 1 Reactions Catalyzed by Citrulline-NO-Cycle Enzymes 3 Table 2 Primers Used for Real Time PCR 69 Table 3 Primers Used for Site Directed Mutagenesis 106 Table 4 Predicted AS serine/th reonine phosphorylation sites 111 Table 5 Possible Sites Phosphorylated by PKA or PKC 120 Table 6 Putative Kinases for Identified AS Serine/Threonine Phosphorylation Sites 132 Table 7 Hypothesized Biological Significance of AS Phosphorylation 133 Table 8 The Basal Nitric Oxide Metabolome 174 Table 9 Putative AS and eNOS Interacting Partners 175

PAGE 13

vii LIST OF FIGURES Figure 1 Working Model of the Regula tion of the Citrulline-NO Cycle Under Physiological Conditions 59 Figure 2 AS Expression Vector Map 65 Figure 3 AS Expression Vector Sequence 66 Figure 4 AS Overexpression Enha nces Nitric Oxide Production 73 Figure 5 Insulin Increases AS a nd eNOS Protein Expression 76 Figure 6 Insulin Increases AS and eNOS mRNA Expression 78 Figure 7 Insulin Enhances AS Promot er Activity at a Distal Element 79 Figure 8 Insulin Enhances Stimulated and Basal NO Production 81 Figure 9 VEGF Increases Endothelial Nitr ic Oxide Production and Upregulates AS and eNOS Expression 83 Figure 10 Ceramide Diminishes AS a nd eNOS Expression and Suppresses eNOS Signaling 85 Figure 11 SDS-PAGE Demons trating AS purification 104 Figure 12 AS is an Endogenous Phosphoprotein, Part I 112 Figure 13 AS is an Endogenous Phosphoprotein, Part II 115 Figure 14 AS is Phosphorylated by PKC and PKA 117 Figure 15 Dose Dependence of AS Phosphorylation 119

PAGE 14

viii Figure 16 VEGF is a Candidate Pathway for Regulating AS Phosphorylation 122 Figure 17 Overexpression and Purificati on of AS Utilizing Immunoprecipitation 125 Figure 18 Overexpression and Purifi cation of AS Utilizing Ni-NTA 127 Figure 19 Identification of AS Ph osphorylation Sites Utilizing Liquid Chromatography-Tandem Mass Spectrometry 129 Figure 20 Multiple sequence alignmen t of AS Phosphorylation Sites 131 Figure 21 The AS 3-Dimensional Structure 134 Figure 22 Structure-Function Relations hips of AS Phosphorylation Sites 135 Figure 23 Role of T131, S180 and S189 on Endothelial Nitric Oxide Production 137 Figure 24 Close Up of Human AS Active Site 141 Figure 25 In Silico Modeling of AS Residues with Good Accessibility for Modification by Phosphorylation 142 Figure 26 Close-Up View of S131, S134 and S328 143 Figure 27 AS Colocalizes w ith eNOS and Caveolin-1 169 Figure 28 AS Co-Immunoprecipitates with HSP90 and Caveolin-1 170 Figure 29 AS Colocalizes with HSP90 171 Figure 30 AS has a Caveolin Binding Motif 172 Figure A-1 The PPAR agonist, troglitazone, stimula tes endothelial NO production and AS protein expression 225 Figure A-2 Troglitazone induces transcription of AS mRNA 226 Figure A-3 Troglitazone induces a distal element in the AS promoter 227 Figure A-4 Troglitazone increa ses binding to the AS PPRE 228

PAGE 15

ix LIST OF ABBREVIATIONS Akt RAC-alpha serine/threonineprotein kinase; Pr otein Kinase B AL Argininosuccinate Lyase AMPK Adenosine Monophosphate Activated Protein Kinase ARP Argininosuccinate S ynthase Regulatory Protein AS Argininosuccinate Synthase ATP Adenosine Triphosphate BAEC Bovine Aortic Endothelial Cells BH4 Tetrahydrobopterin CKII Casein Kinase II eNOS Endothelial Nitric Oxide Synthase FAD Flavin Adenine Dinucleotide FMN Flavin Mononucleotide GSK3 Glycogen Synthase Kinase 3 Beta HSP90 Heat Shock Protein 90 IFN Interferon Gamma iNOS Inducible Nitric Oxide Synthase IP Immunoprecipitation LC-MS/MS Liquid Chromatography-Tandem Mass Spectrometry

PAGE 16

x LPS Lipopolysaccharide MDLA -Methyl-DL-Aspartic Acid NADP Nicotinamide Adenine Dinucleotide Phosphate Ni-NTA Nickel Nitrilotriacetic Acid nNOS Neuronal Nitric Oxide Synthase NO Nitric Oxide NOS Nitric Oxide Synthase PKA Protein Kinase A PKC Protein Kinase C PKG Protein Kinase G PPAR Peroxisome Proliferator Activated Receptor PTM Post-Translational Modifications ROS Reactive Oxygen Species STZ Streptozotocin TNF Tumor Necrosis Factor Alpha TZD Thiazolidenedione VEGF Vascular Endot helial Growth Factor VSMC Vascular Smooth Muscle Cells

PAGE 17

xi MULTI-LEVEL REGULATION OF EN DOTHELIAL ARGININOSUCCINATE SYNTHASE: SIGNIFICANCE FOR NITRIC OXIDE PRODUCTION Karen Davidowitz Corbin, MS, RD ABSTRACT The citrulline-nitric oxide (NO) cycle, comprised of the enzymes argininosuccinate synthase (A S), argininosuccinate lyase (AL) and endothelial nitric oxide synthase (eNOS), is responsible for the regulated production of endothelial NO. Although most studies have focused on e NOS to uncover important regulatory mechanisms, we and others have determined that AS is an essential and regulated step in endothelial NO production. AS is rate limiting for endothelial NO production and is the primary source of arginine, the substrate for eNOS-mediated NO production, despite saturating intracellular levels of arginine and available arginine transport systems. AS is essential for endothelial cell vi ability and its expression is regulated coordinately with eNOS by TNF and thiazolidenediones with concomitant effects on NO production. Given the importance of AS for endothelial health, we explored three independent regulatory mechanisms. In Chapter One, th e functional consequen ces of altered AS expression due to overexpression, insulin, VEGF and ceramide were studied. We

PAGE 18

xii demonstrated that overexpression of AS leads to enhanced NO production and that insulin, VEGF and ceramide coordinately re gulate the expression of AS and eNOS. In Chapter Two, the first post-translational modifications of AS in the endothelium were characterized. We determined that AS is an endogenous phosphoprotein in the endothelium, described several levels of bi ological significance of AS phosphorylation, identified 7 sites of AS phosphorylation a nd began to uncover the direct impact of phosphorylation on AS function. Finally, in Chap ter Three, endothelial AS subcellular localization was defined and important protei n interactions were identified including caveolin-1 and HSP90. The work presented in th is dissertation demonstrates that multiple mechanisms regulate the function of AS, often coordinately with eNOS, and have a direct impact on nitric oxide production. Our findings suggest that the globa l understanding of the citrulline-NO cycle as a metabolic unit will unravel new paradigms that will re-define our understanding of th e regulation of vascular function by NO.

PAGE 19

1 INTRODUCTION Nitric Oxide: Historical Perspectives and Biochemistry Nitric Oxide (abbreviation: NO; chemical formula: N O) is a readily diffusing gas that can be poisonous in the environmen t yet powerful in biology [1, 2]. Due to its unpaired electron, NO is a free radical, making it highly reactive. The Molecule of the Year in 1992 [3, 4], the discoveries made about NO and heart function earned Robert F. Furchgott, Louis J. Ignarro and Ferid Murad the Nobel Prize in 1998 (NobelPrize.org). In 1980, endothelium-dependent vessel relaxation wa s first described [5] and in 1987, it was determined that the endothelium derived rela xing factor (EDRF) was in fact NO [6, 7]. NO moves easily in and out of cells so it cannot be stored inside producing cells like other endogenous messengers. In the presen ce of oxygen, it has a ha lf life of just a few seconds, but its longevity in the body is not known [1, 2]. There are several nitrogenderived compounds that exert di stinct biological functions including NO+ (nitrosonium), NO (nitric oxide) and NO(nitroxyl anion). Most stud ies of nitric oxide do not differentiate or define the specific species that is exerting biological ac tions [2]. From this point forward, our discussions will be related to NO in general terms.

PAGE 20

2 NO functions via multiple mechanisms to exert biological effects. It reacts directly or indirectly with proteins, lipids, nucleic acids, metals, other gases (such as oxygen) and carbohydrates [1]. Like many othe r molecules, the effects of NO can be either positive or negative depending on the cellular environment and amounts produced. Both a lack of NO and an excess of NO can l ead to pathological consequences such as hypertension and septic shock, respectively [3 ]. NO is a signaling agent and has been implicated in a broad number of functi ons including neurotra nsmission, memory, host defense, vasodilation, blood flow, respiration, nutrient metabolism and apoptosis [3, 810]. Therefore, the chemical simplicity of this molecule has no bearing on its broad and essential functions. The Citrulline-Nitric Oxide Cycle The citrulline-NO cycle, comprised of the enzymes argininosuccinate synthase (AS), argininosuccinate lyase (AL) and endothelial nitric oxide synthase (eNOS), is responsible for the regulated production of nitric oxide [ 11]. AS is a homotetrametic enzyme with a molecular weight of ~47 kD a per monomer. Each monomer contains 412 amino acids. It is transcribed from a single gene on chromosome 9 and is expressed in virtually all tissues with the highest expre ssion in liver, kidney and brain [12-17]. AL is also a homotetramer with a subunit molecular we ight of ~50 kDa. The human AL gene is found on chromosome 7 and has 464 amino acids per monomer. It is expressed widely, similar to AS [13, 17, 18]. There are 3 isofor ms of nitric oxide synthase (NOS): neuronal NOS (nNOS; NOS1), inducible NOS (iNOS ; NOS2) and endothelial NOS (eNOS;

PAGE 21

NOS3). Each is encoded by a separate ge ne on chromosomes 12, 17 and 7, respectively, and they share ~ 50-60% sequence homology [19, 20]. eNOS, the NOS isoform studied in this work, is a homodimer with a molecular weight of 135 kDa per subunit that is expressed in endothelium, skeletal and cardiac muscles, kidney tubules and many other non-endothelial tissues [21, 22]. eNOS, like nNOS is a constitutive NOS isoform that is regulated by calcium/calmodulin. Th is is in contrast to iNOS which is active even at low calcium concentrations [19]. Besides the substr ates necessary to carry out its reaction (see Table 1), eNOS requires tetrahydrobiopterin (BH4), flavin adenin e dinucleotide (FAD), flavin mononucleotide (FMN) and heme as co-factors. The amounts of NO produced by constitutive isoforms is ~ 1000 fold less than iNOS [19, 20]. The following are the reactions cataly zed by each of the enzymes in the endothelial citrulline-NO cycle: Table 1: Reactions Catalyzed by Citrulline-NO-Cycle Enzymes. eNOS (EC 1.14.13.39): L-arginine + NADPH + H + O2 = L-citrulline + NO + NADP+ AS (EC 6.3.4.5): (rate limiting step) ATP + L-citrulline + L-aspartate = AMP + diphosphate + argininosuccinate AL (EC 4.3.2.1): argininosuccinate = furmarate + L-arginine 3

PAGE 22

4 The function of the citrulline-NO cycle as a whole is crucial for the production of NO. The availability of argini ne is a key factor limiting NO synthesis, despite the fact that intracellular concentrat ions of arginine (0.1-0.8 mM) are greatly in excess of the reported eNOS Km (~5 M) [11]. This phenomenon is termed the arginine paradox. Although there are transport systems available that have been hypothe sized to supply the arginine utilized for NO production [23], we a nd others have shown that endothelial NO production is limited by the capacity to regene rate arginine from citrulline [24-27]. In fact, cells that express cons titutive NOS isoforms, such as endothelial cells and neurons have the capacity to regenerate arginine from citrulline [25, 28]. Dysfunction of AS and AL, which are also part of the urea cycle in ureagenic tissues, can lead to metabolic defects termed collectively as urea cycle disorders, since deficiency in any one of the six enzymes in the urea cycle causes disease [29]. Type I citrullinemia is an autosomal recessive disorder caused by AS deficiency (OMIM #215700) [30]. Some of the symp toms of this disease include severe vomiting, excess levels of citrulline in serum, spinal fluid and urine, hyperammonemia and mental retardation. If not treated promptly, it can lead to death. Death in the neonatal period occurs in nearly half of cases [29]. Many mu tations have been identified in the AS gene that can lead to type I citrullinemia [31, 32]. Unlike classical citrullinemia in children that results from a mutation in the AS gene and is associated with an overall defici ency of AS, in type II citrullinemia there is no mutation in the AS gene (OMIM #603471) [33]. AS protein has normal kinetic

PAGE 23

5 properties and is quantitatively deficient only in the liver. The gene that is actually defective is SLC25A13. It encodes a mitochondrial Ca2+-dependent aspartate/glutamate transporter called citrin [33, 34]. The loss of organization attribut able to the mutated citrin leads to reduction of AS protei n, possibly through destabilization and/or degradation. Symptoms of t ype II citrullinemia include enuresis, delayed menarche, insomnia, nocturnal sweats, recurrent vomiting, diarrhea, tremors, episodes of confusion after meals, lethargy, convulsions, delusions, ha llucinations, and brief episodes of coma. Some patients die within a few years of onset [29]. For both type I and type II citrullinemia, a low protein diet, medicatio ns to reduce ammonia levels (examples: sodium benzoate, sodium phenylacetate) or dialysis can be used as treatments to reduce ammonia levels [35]. Argininosuccinic aciduria is an autosomal recessive disorder caused by multiple possible mutations in the AL gene (O MIM 207900) [36]. Like citrullinemia, argininosuccinic aciduria can be early onset (more severe) or late onset. Some symptoms of early onset AL deficiency include leth argy, skin lesions, mental retardation, liver enlargement and convulsions. Late onset disease is usually mild and symptoms tend to occur during illness or periods of stress. Treatment for argininosuccinic aciduria is the same as for citrullinemia [35].

PAGE 24

6 Endothelial Nitric Oxide and Vascular Health NO regulates most normal functions of the endothelium. One primary function of NO is vasodilation. This process occurs when NO produced in the endothelium migrates to the adjacent smooth muscle layer, binds to the heme moiety of soluble guanylyl cyclase (sGC) and activates it. This lead s to an increase in cyclic guanosine monophosphate (cGMP), which then activates protein kinase G (P KG) and leads to vasodilation. Regulation of vessel dilation co ntributes to blood fl ow and blood pressure control [37, 38]. In addition, NO regulates plat elet aggregation. The health of arteries is highly dependent on tight control over th is process. During injury, plat elets migrate to the site of injury and form a plug to seal the blood vessel and minimize blood loss. On the other hand, uncontrolled platelet aggregation diminishes the fluidity of blood, prevents the delivery of oxygen and nutrients to tissues and can lead to thrombosisa leading event in myocardial infarction. During physiological conditions, the balance is tipped towards reducing platelet aggregation. Both endothelial and platel et derived NO lead to the activation of PKG, the inhibition of cA MP phosphodiesterase and the reduction of cytosolic calcium. This leads to decreased platelet aggregation and adhesion and a disaggregation of existing plat elet aggregates [39, 40]. Another key vascular function regulated by NO is angiogenesis, which is defined as the formation of new blood vessels. If unchecked, this process can be pathogenic.

PAGE 25

7 Physiological angiogenesis is important for improving ischemic conditions and for wound healing [41, 42]. Vasodila tion is required for angiogenesis as evidenced by the fact that many angiogenic factors, such as VEGF, possess vasodilating properties [43, 44]. NO has been demonstrated to have a direct role in angiogenesis since NO donors promote and NO inhibitors diminish angioge nesis [45-48]. NO promotes angiogenesis via cGMP elevation and by e nhancing the expression of angiogenic factors [41, 42, 48, 49]. Enhancement of NO pr oduction in patients undergoi ng coronary artery bypass grafting by administration of l-arginine in conjunction with VEGF improved angiogenesis and surgical outcomes, thus ex emplifying the tight association between NO signaling and angiogenesis [50]. NO can be both pro-and anti apoptotic. High concentrations of NO induce cell death in macrophages, pancreatic islet cells tumors and other cell types. There are several mechanisms by which NO can promot e apoptosis. First, NO can activate the mitochondrial apoptosis pathway by increasing the release of cytochrome c. This leads to activation of the caspase-dependent apoptos is pathway. NO can also cause DNA damage which induces p53 and leads to cell cycle arrest. Excess NO activ ates pro-apoptotic kinase cascades such as c-Jun N-terminal ki nase and mitogen activated protein kinases. Finally, NO can cause an increase in the releas e of ceramide, a bioactive sphingolipid that can itself induce apoptotic pathways via the death receptor pathway [51-54]. Cytokines, high glucose lipids mediate some of the proangiogenic properties of NO [55-60]. NO deficiency also leads to apoptosis [51-54].

PAGE 26

8 On the other hand, physiological concen trations of NO protect cells from apoptosis [51-54]. One mechanism of protect ion is via the rel ease of cGMP, which subsequently diminishes cytochrome c release, caspase activation and ceramide accumulation. In addition, NO can lead to suppr ession of caspase activity by reversible Snitrosylation and increase the expression of anti-apoptotic genes such as Bcl-2 and HSP70 [54]. Shear stress is an important m echanism that protects endothelial cells from TNF or growth factor withdrawal-induced apoptosis [61]. One mechanism proposed for this protective effect is via the upregulation of NOS a nd superoxide dismutase [62]. Considering the plethora of vascul ar functions regulated by NO, any dysregulation of its bioavailabi lity and production has far re aching consequences. In fact, endothelial dysfunction is asso ciated with a number of di sorders including diabetes, obesity, atherosclerosis, meta bolic syndrome and hypertension [63-67]. Much of what is known related to endothelial dysfunction and impaired NO production is related to the expression, activation, localization, interactions and substrate/cofactor availability of eNOS [63]. The role of other citrullin e-NO cycle components remains relatively unexplored. Endothelial Dysfunction There are a number of mechanisms that have been implicated as contributing events in endothelial dysfunc tion, and most are related to impaired function of NO. One hallmark of endothelial dysfunction is inflammation [68]. NO possesses anti-

PAGE 27

9 inflammatory properties that are disrupted in endothelial dysfunction via mechanisms involving increased leukocyte adhesion, increased expression of chemokines, increased monocyte migration and infiltra tion of monocytes into the ar terial wall [65, 69]. Another inflammatory process disrupted in endothelial dysfunction is related to an increased level of C-reactive protein (CRP). CRP dimini shes the production of NO by reducing eNOS expression [69]. Multiple other inflammatory molecules are elevated in endothelial dysfunction such as TNF Whether elevated TNF is a cause or consequence of endothelial dysfunction is debatable. Howeve r, there are multiple mechanisms by which TNF causes endothelial dysfunction such as th e inhibition of insulin signaling, an increase in lipids such as ceram ide and increased apoptosis [70-77]. In endothelial dysfunction, there is also an increase in reactive oxygen species (ROS). This is one central mechanism by whic h cardiovascular disease risk factors such as diabetes, hypercholesterolemia and hyperten sion lead to endothelial dysfunction [63]. One such species, superoxide, reacts w ith NO and leads to the production of peroxynitrite, a toxic free radical that diminishes the bioava ilability of NO and inhibits the protective actions of NO [78]. ROS also degrade BH4, an important eNOS co-factor. This leads to the uncoupling of NADPH oxi dation from eNOS-mediated NO synthesis and the subsequent production of su peroxide instead of NO [79-81]. Another consequence of increased ROS is the inhibition of dimethylarginine dimethylaminohydrolase (DDAH). This leads to increased levels of asymmetric dimethylarginine (ADMA), an endogenous e NOS inhibitor [82]. ADMA inhibits eNOS

PAGE 28

10 by multiple mechanisms such as further in creasing superoxide production and eNOS uncoupling, diminishing the intera ction of HSP90 with eNOS and redistributing eNOS to the mitochondria with subsequent mitoc hondrial dysfunction [ 83-86]. ADMA has been linked to vascular disease by contributing to reperfusi on injury and increasing blood pressure [84, 87]. Protective effects at early time points have also been shown via iNOS mediated inhibition of intima-media thickness [88]. Elevated lipids also influence the func tion of the endothelium. For example, oxidized LDL reduces eNOS activity by disp lacing it from the plasma membrane and targeting to intracellular sites where it is less active [89]. Elevated ceramide levels due to excess saturated fat in the diet or chroni c inflammation diminish eNOS activation and NO production [60, 90, 91]. Another lipid linked to endothelial dysfunction is apolipoprotein CIII (apoCIII), a component of very low density and low density lipoprotein that is elevated in insulin re sistance and metabolic syndrome. ApoCIII activates pro-inflammatory atherogenic pathways by directly inhibi ting insulin signaling and reducing the activation of e NOS [92]. In fact, one of the benefits of statin therapy beyond cholesterol lowering is the improved fu nction of eNOS and endothelial function. One example is improved blood flow to ischemic tissues in diabetes in response to statins due to an upregulation of eNOS [93]. Ther e are controversies about whether statins improve endothelial function and this is perhap s associated with other factors associated with endothelial dysfunction. For instance, simvastatin is able to improve endotheliumdependent vasodilation in patients with low le vels of ADMA but is in effective in patients

PAGE 29

11 with high ADMA. On the other hand, argini ne supplementation in combination with statins works well with patients with high but not low ADMA [94]. Insulin resistance and diabetes are also tightly associated with endothelial dysfunction [73]. In fact, macrovascular complications are quite common in diabetic patients leading to the classifi cation of diabetes as a vascul ar disorder [95]. Endothelial cells express insulin receptors and many cardioprotective signaling pathways are mediated by insulin [95]. Several derangeme nts can occur when insulin function is compromised including an increase in ROS, depletion of BH4, lipid abnormalities and hypertension [96-100]. Hyperglycemia, a conse quence of diminished insulin action, also has direct impacts on eNOS expression and activity and enhances endothelial cell apoptosis [58, 101, 102]. Collect ively, these imbalances impair the ability of the endothelium to produce adequate amounts of NO and all norm al endothelial functions are compromised leading to a stri king association between insuli n resistance, diabetes and coronary artery disease. Overall, the mechanisms that lead to endothelial dysfunc tion are numerous and complex. Most of them converge and lead to one major issuea decrease in the production and bioavailability of NO. Therefore, a better understanding of the mechanisms that control NO production is e ssential for the development of effective prevention and treatment strategi es for vascular disorders.

PAGE 30

12 Prevention of Vascular Disorders There are multiple medical and surgical approaches to treat cardiovascular diseases and they should be employed as early as possible to diminish long-term damage. Considering the far reaching consequences of endothelial dysfunction and the fact that by the time significant symptoms arise, va scular disease has already progressed dramatically, prevention strate gies are the best insurance against the development of vascular disease [67]. Since most risk factors for heart disease mediate their deleterious effects by impairing NO production [63], preventa ble risk factors need to be addressed either with lifestyle modi fication of medical interven tions. Healthy eating and maintaining a normal body weight can go an extremely long way towards keeping blood pressure within normal limits, preventing diabetes and metabolic syndrome and maintaining a healthy level of blood lipids [103, 104]. Cigarette smoking directly impairs NO bioavailability [105], so avoiding smoking or quitting is essentia l. In addition, blood pressure and cholesterol, if inadequately responsive to lifestyle measures, should be treated pharmacologically [106]. Argininosuccinate Synthase Functions Related to Nitric Oxide Production AS, the focus of this dissertation, has se veral unique properties that are essential for NO production and vascular health. AS is rate limiting for the production of NO [107]. This was initially demonstrated in vascular smooth muscle cells when AS overexpression enhanced the cap acity of transfected cells to produce NO, despite non-

PAGE 31

13 limiting levels of arginine [108]. In endothe lial cells, we have demonstrated that inhibition of AS activity with siRNA or the AS inhibitor -methyl-DL-aspartic acid diminishes NO production due to reduced AS activity [24, 109]. We also demonstrated that the cofractionation of AS, AL and eNOS in plasmalemmal caveolae allowed endothelial cells to effectively distinguish bulk intracellular arginine from the arginine used for NO production [110]. Our substrate utilization studies supported this view by demonstrating that the r ecycling of citrulline back to arginine is an efficient, tightly c oupled process [24]. The ac tivity of AS in this recycling process is the preferred mech anism for eNOS-mediated NO production, while iNOS-mediated NO production in smooth muscle is dependent on arginine transport [26]. One additional and unique mechanism of regulation of NO production by AS involves expression of endothelium -specific AS variants with different lengths of the 5 untranslated region (5 UTR). Th e shortest form of the message represents ~90% of the total AS mRNA and encodes for full length AS. The two longer forms contain an out of frame upstream open reading frame (uORF) that encodes a 4 kDa protein called Argininosuccinate Synthase Re gulatory Protein (ARP). Over expression of ARP leads to diminished NO production and siRNA knockdown of ARP leads to an increase in NO production in endothelial cells [111]. The three dimensional structure of human AS has revealed an interesting feature relevant to its role in NO produc tion. It has been demonstrated that nitrosylation of AS at

PAGE 32

14 cysteine 132 (C132) leads to inactivation of the enzyme [112]. A mechanism was not identified for this activation. The inves tigators who solved th e human AS crystal structure used in silico modeling to predict the effect of nitrosylation at this site on the catalytic efficiency of AS. The authors not ed that C132 is actua lly buried in the Nterminal domain. They suggested that a signifi cant structural change would have to take place for the nitrosylation to occur. Based on their in silico models, they determined that even though C132 is not near the active site, the modification does perturb the orientation of amino acids that are part of the active si te [113]. This presents the first evidence of a possible conformational change caused by a modification at an allosteric site which then changes the activity of AS with s ubsequent effects on NO synthesis. Regulation of Endothelial Nitric Oxide Production: Protein Expression The functions of NO are cont rolled at the levels of bi osynthesis and availability because it cannot be stored. Since virtuall y all phenotypic properties of normally functioning endothelial cells ar e related to NO, its production is tightly controlled [28, 114, 115]. This tight control requires multip le regulatory mechanisms. One such mechanism is the regulation of the expression of the enzymes of the citrulline-NO cycle. To assess the importance of the expression citrulline-NO cycle components on vascular function, animal knockout models have been generated and have yielded important information. The eNOS knockout mouse models have by far been the most prominent in delineating the importance of the citrulline-NO cycle for vascular health.

PAGE 33

15 Deletion of the eNOS gene is not embr yonic lethal, but does lead to hypertension, impaired wound healing and angiogenesis [ 116]. In addition, eNOS knockout mice have hyperinsulinemia, hyperlipidemia, diminished glucose metabolism [117] and a decrease in skeletal muscle oxidative capacity [118]. In contrast, eNOS overexpression ameliorates vascular dysfunction by improvements in lipid s, blood pressure, myocardial contraction and protection against ischemia -reperfusion injury [119, 120] The lessons we have learned from eN OS knockout mice are complicated by the fact that in some instances, a lack of eNOS has some beneficial effects and overexpression of eNOS has deleterious eff ects. For example, eNOS knockout mice have decreased formation of diet-induced fatty streaks [121] and endothelium dependent relaxation and bradykinin-mediated blood flow are intact due to a compensatory upregulation of nNOS [122]. Although physiolo gical levels of NO are beneficial for myocardial contractility, overe xpression leads to a negative in otropic effect [123]. Due to the difficulty in interpreting mouse data rela ted to compensatory functions of the NOS isoforms, a triple knockout mouse where eNOS, nNOS and iNOS were deleted was recently generated. These mice had a much lower survival rate and displayed a long list of metabolic abnormalities including meta bolic syndrome, obesity, hyperlipidemia, hypertension, and impaired glucose tolerance [124]. Taken together, animal models of reduced or enhanced eNOS expression dem onstrate the essential role of eNOS in mediating vascular health.

PAGE 34

16 In contrast, when AS is knocked out in mice, it leads to death a few days after birth due to the accumulation of citrulline a nd a severe arginine deficiency [29, 125]. A tissue specific AS knockout mouse model is not available, but if, for example, such a model were generated with an endothelium specific deletion of AS, we would predict marked endothelial dysfunction based on our tissue culture work. Similarly to the AS knockout, AL knockout mice die within 48 hour s of birth and are characterized by increased ammonia, argininosuccinic acid, glut amine and citrulline and low levels of arginine [29, 36]. The role of the expression citrullineNO cycle components on vascular health can be studied at extreme levels ut ilizing animal models as just described. These scenarios are rare in humans, and there are finer levels of regulation that occur in response to a number of biological processes and signals. Shear stress is one of the most important physiological factors that regulates the function of bl ood vessels [126-128]. Unidirectional, laminar shear stress exerts a protective role on the endothelium by diminishing adhesion and platelet aggregation [129] a nd enhancing NO production [126, 130]. Shear stress has been shown to increase eNOS expression via a transient activation of Nf B in bovine aortic endotheli al cells. There is a shear stress responsive element (SSRE) in the eNOS promoter that is res ponsible for the transc riptional regulation by Nf B [131]. Other groups have determined that the shear stress-mediated increase in eNOS mRNA is due to an increase in transc ription and mRNA stabil ization [132, 133]. In a DNA microarray study of human umbilical ve in endothelial cells, shear stress was shown to significantly increase AS expressi on, but not eNOS. The authors suggested this

PAGE 35

17 might be due to the fact that shear stre ss-induced increases in NO synthesis depend on arginine synthesis from citrulli ne via increased AS levels [134]. In a subsequent study by a group studying the adhesion of endothelial cells in vein grafts supported the role of both eNOS and AS in the response of the endothelium to shear stress [135]. The role of AL in shear stress-mediated NO production has not been studied. However, it is clear that there is a role for both AS and eNOS in mediati ng the response of the endothelium to shear stress that involves multiple and perhaps overlapping mechanisms depending on the level of shear stress and other factors. Cytokines regulate vascular function in various ways, and tumor necrosis factor alpha (TNF ) is well characterized in this respect. TNF is a pro-inflammatory cytokine with both positive and negativ e roles in regulating vasc ular function, depending on the length of treatment and tissue type [136]. In general, conditio ns of chronic TNF elevations are deleterious to the vasculature as evidenced by the fact that humans with diabetes, obesity and heart disease have elevated levels of TNF [55, 71, 73, 75, 137, 138]. Often times, TNF -mediated vascular impairment is caused by dysregulation of NO production [70, 74, 139, 140]. One mechanism by which TNF inhibits NO production is by altering eNOS and AS expression. Specifi cally, eNOS expression has been shown to be downregulated by TNF via several mechanisms including decreased mR NA stability and inhibition of transcription fact or binding [141-143]. In addition, TNF diminishes AS expression and NO production by in hibiting the binding of SP-1 elements to the proximal promoter [144]. AL seems to be less amenable to cytokine regulation since in vascular smooth muscle cells, retinal endothelium or pancreatic cells, treatment with a variety

PAGE 36

18 of different cytokines led to co-induction of AS and iNOS while AL remained static [145-147]. Peroxisome proliferator-ac tivated receptor gamma (PPAR ) is a member of the nuclear receptor superfamily of ligand-activated transcription factors. PPAR regulates the transcription of genes i nvolved in lipid and glucose me tabolism [148], differentiation [148-150] and cell growth [148] Thiazolidinediones (TZDs) are a group of synthetic PPAR agonists that provide card iovascular benefits such as reduction of blood pressure and improvement in insulin-sensitivity [151-153]. In addition, TZDs reduce lesion formation in animal models of atherosclerosis [154156], improve flow-mediated vasodilation and decrease vascular smoot h muscle cell migra tion by stimulating endothelial NO production [157, 158]. These anti-diabetic compounds also counter the effects of the inflammatory response associ ated with elevated serum levels of TNF and ceramide. For instance, ro siglitazone impairs TNF -induced activation of MAPK, restoring insulin signaling and leading to normalization of glucose uptake in brown adipocytes [159]. In humans with type 2 di abetes, pioglitazone was protective against TNF-mediated endothelial dysf unction [76]. In hepatocytes made insulin resistant via TNF treatment, troglitazone is able to restor e insulin sensitivity, partially via blocking the downstream actions of ceramide [160]. Another mechanism by which TZDs improve endothelial func tion is by inducing the expression of components of the citrulli ne-NO cycle. For example, eNOS expression and NO production are enhanced by troglitazone by a mechanism that seems to be PPAR

PAGE 37

19 independent [161]. Telmisartan, an angiotensi n II receptor blocker, has recently been identified as a partial PPAR agonist. Its cardioprotectiv e mechanism is mediated partially by the upregulation of eNOS [153] In addition, troglitazone enhances NO production and AS expression by activating a PP AR responsive element in the distal AS promoter [162]. There have been no studies on the effects of PPAR agonists on AL. Taken together, the positive benefits of TZDs are multifactorial and lead to improved vascular parameters by mechanisms that in clude the upregulation of eNOS and AS. Insulin is an important regulator of vasc ular function. Insulin mediates its effects on vasodilation by stimulating NO release [ 163, 164]. One mechanism by which insulin increases NO production is by increa sing the expression of eNOS. Kuboki et al. demonstrated that insulin increases eNOS mRNA within 1 hour of treatment. This increase was mediated by activation of the PI3-kinase pathway [165]. In another study, the mechanism of insulin transcriptional regulation of eNOS was found to involve increased binding of SP-1 and AP-1 transcript ion factors [166]. The link between insulin and NO has another important feature. NO itsel f leads to the secretion of insulin in pancreatic islet -cells [167]. This process is dependent on the recycling of arginine to citrulline [168]. This suggests an involveme nt of both AS and AL. In addition, eNOS expression and function is diminished in diab etes. For example, high glucose diminishes the expression of eNOS [102]. In addition, insulin enhances the expression of eNOS in diabetic rats [169]. The impact of insulin or diabetes on endothelial AS and AL expression has not been studi ed, but one group did note that in the early phases of streptozotocin (STZ)-induced type 1 diabetes in rats, AS and eNOS mRNA and protein

PAGE 38

20 expression was induced in aorta. This increas e then diminished as diabetes progressed. There was very little effect on AL mRNA, a nd possibly a slight induction after 4 weeks, but the effects on AL were not studied any fu rther [170]. Thus, glucose, insulin and NO metabolism are tightly associated processes th at involve both AS and eNOS (and possibly AL) expression. VEGF is an important mediator of vasc ular health by regula ting vasodilation and angiogenesis [171]. VEGF exerts its cardioprote ctive functions, in part, by increasing the expression of eNOS. In human umbilical vein endothelial cells (HUVECs), VEGF augments the expression of eNOS in a tim e and dose-dependent manner. This was linked to increased basal and stimulated NO produc tion [172]. Another group obtained similar results when using rat aortic rings [173]. In addition, insuli n increases the expression of both eNOS and VEGF in diabetic rats lead ing to increased NO production and vascular relaxation [169]. The role of VEGF in re gulating AS or AL has not been studied. Therefore, VEGF and NO signaling pathways are linked by dire ct effects on eNOS expression. Ceramide, a bioactive sphingolipid, has been implicated in the development of insulin resistance and atherosclerosis [57, 61, 77, 174]. In general, supraphysioloical levels of ceramide contribute to endothelial dysfunction [60, 91, 175] and most of what is known involves impairment of e NOS activation (which is discussed in the next section). However, Li and colleagues demonstrated that ceramide decreases NO production in human endothelial cells via a mechanism th at involves ROS genera tion. They found that

PAGE 39

21 eNOS expression is enhanced as a compensato ry mechanism. This enhanced expression was not sufficient to ameliorate the dimini shed NO synthesis caused by the generation of ROS [90]. The direct effect s of ceramide on AS and AL expression have not been studied. The intricate and diverse mechanisms re gulating the expression of AS and eNOS in the endothelium and other tissues suggests that this is an important mechanism mediating vascular biology and warrants con tinued investigation. For this reason, the regulation of AS expression in endothelial cells encompasses Specific Aim 1 of this dissertation and is described in Chapter One. Regulation of Endothelial Nitric Oxide Pr oduction: Post-Translational Modifications Although controlling the expression levels of AS, AL and eNOS is an important mechanism for regulating the levels of NO produced, this mode of regulation seldom occurs quickly enough to acutely increase or decrease NO production. Since NO is constantly being produced and its levels are fr equently adjusted to meet cellular demands, post-translational modifications (PTM) have b een described as a prominent mechanism in regulating NO production. The acute regulation of eNOS by reve rsible phosphorylation is well described. Although eNOS is regulated by tyrosine phos phorylation [176, 177], serine/threonine phosphorylation is a much more prominent me chanism. There are 5 well characterized

PAGE 40

22 sites of eNOS phosphorylation: serine 116 (S116), threonine 497 (T497), S617, S635 and S1179 (bovine; these sites are equivalent to S114, T495, S633 and S1177 in human) [178-180]. S116 is located in the oxygenase doma in of the enzyme [179]. There is controversy as to whether this site acti vates or inactivates eNOS. A recent paper demonstrated that mutation of this site to an alanine increased NO production while mutation to an aspartic acid diminished NO production. In addition, phosphorylation at this site increased the inhibitory interacti on of eNOS with caveo lin-1 and diminished vascular reactivity. Hence, this site seems to be inhibitory for the function of eNOS [181]. On the other hand, Drew et al found that HDL and Apo A1 led to increased NO release due to eNOS phosphor ylation at several sites, most profoundly at S116. This suggested a positive role for S116 [182]. It is possible that there are tissue and stimulusspecific functions for S116. There is a great deal of consensus regard ing the inhibitory ro le of T497, which is located in the eNOS calmodulin binding domai n [179]. In fact, in the study mentioned above where they found that S116 does activate eNOS, they also found that HDL and Apo A1 diminish phosphorylation at T497 [182]. Many other studies have confirmed this inhibitory regulation [81, 183-185]. In particular, se veral different groups have demonstrated that mutation of T497 to aspartic acid (phospho-mimetic) greatly diminishes eNOS activity and NO production while the alanine mutation (phospho-null) has opposite effects [79, 184]. When this s ite is phosphorylated, calmodulin cannot bind

PAGE 41

23 to and activate eNOS [184]. T497 is constitutively phosphorylated and upon stimulation with agents that increase intr acellular calcium, such as bradykinin or VEGF, the site is dephosphorylated by protein phosphatase 1 (PP1) leading to calmodulin binding [184]. In addition, the dephosphorylation of T497 ha s to occur before the activating phosphorylation at S1179 can take place [185]. Protein kinase C has been shown to increase phosphorylation at this site [183] thereby diminishing eNOS function. Certain PKC isoforms, such as PKC and PKC can activate eNOS by leading to phosphorylation at S1179 [186, 187]. Serines 617 and 635 are both in the eNOS auto inhibitory domain, which is part of the reductase domain [179]. This domain is repor ted to keep eNOS in an inhibited state that is reversed upon calmodulin binding [ 19, 188]. These sites are phosphorylated in response to bradykinin, ATP and VEGF [179, 182, 189]. Shear stress has also been found to lead to the phosphorylation of S635 [190]. The phosphorylation at S617 is transient in nature while S635 phosphorylation is more persistent. Akt phosphorylates S617 while PKA phosphorylates S635. The mechanism of regulation of eNOS activation by these two sites is suggested to involve a two st ep process. During in itial agonist-induced activation of eNOS, S617 is phosphorylated and increases eNOS sensitivity to calcium. Then, S635 phosphorylation increases eNOS specific activity to levels similar to what occurs when S1179 is phosphorylated [189]. The most studied site of eNOS phosphor ylation is S1179, lo cated near the Cterminus in the reductase domain [179]. P hosphorylation at this site leads to an

PAGE 42

24 approximately 2-fold activation of eNOS speci fic activity [189]. This activation is due to a decrease in the calcium-dependence of eNOS and an increase in electron flux from the reductase domain to the oxygenase domain, which leads to increased NO output [191]. This site is phosphorylated by a number of kinases including Akt, AMPK, PKA, PKG and calmodulin II protein kinase in response to a variety of stimuli including insulin, adiponectin, bradykinin, VEGF, HDL and Apo A1 [127, 130, 182, 183, 185, 186, 192202]. For example, fluid shear stress, VEGF and bradykinin lead to the activation of PKA and subsequent phosphorylation of eNOS at S1179 [183, 203]. Although the regulation of eNOS by a complex pattern of stimulus-specific phosphorylation and dephosphorylation events ha s been studied extensively, there have been no reports of such studies for AS and AL. Evidence supporting post-translational regulation of AS is beginning to accumulate. First, in a proteomic study to identify novel phosphoproteins in HeLa cells, AS was found to be phosphorylated at S352 [204]. This is the first and only indication of this type of modification of AS. The authors did not investigate the biological relevance of this modification. In addition, in vascular smooth muscle, AS is reversibly inactivated by nitr osylation in conditions of high NO output by iNOS. Their work suggests that AS is at least partially responsible for sensing cellular NO levels and adjusting output accordingly in an effort to maintain homeostasis [112]. Nitrosylation of cysteine residues has emerged as an extremely important, reversible modification that regulates the activity of a number of proteins such as caspase-3, the ryanodine receptor and actin [205]. Importantly, eNOS is basally nitrosylated in endothelial cells and is then rapidly denitrosylated in re sponse to VEGF stimulation

PAGE 43

25 [206]. The nitrosylation of AS in the endothe lium has not been studied, but may represent an important additional m echanism of regulation. Another important PTM that regulates NO production is the acylation of eNOS by the irreversible N-myristoylation at glycine 2 (G2) and the reversible thiopalmitoylation at cysteine 15 (C15) and C26 [207]. These m odifications are crucial for the targeting and anchoring of eNOS to the plasma membrane which is an essential location for the production of NO [207-210]. Thes e modifications have not been studied for AS or AL. However, we have demonstrated that AS and AL and eNOS co-fracti onate with caveolin1 in caveolar membrane fractions, which indicat es that mechanisms must exist to target these enzymes to the plasma membrane in conjunction with eNOS. Glycosylation is an additional type of PTM that regulates NO production. Olinked glycosylation occurs at serine and threonine residues, typically in response to the flux of excess glucose into the hexosamine biosynthesis pathway. Glycosylation often competes with phosphorylation at the same or adjacent sites leading to a ying-yang type of regulatory scheme [211, 212]. In endothelial cells, glycosyl ation of eNOS decreases its activity and diminishes phosphorylation at S1179 [213]. The functional relevance of this modification was illustrated in diabetes-related erectile dysfunction, where glycosylation reduced S1179 phosphorylation at baseline and in response to VEGF and shear stress [214]. In this study, it was determined th at glycosylation only affects S1179 and not T497, S617 or S635. Although glycosylation of AS in endothelial cells has not been studied, in Caco-2 cells the expression of AS is stimulated by glutamine via glycosylation

PAGE 44

26 of the transcription factor SP-1 [215]. This modification leads to nuclear import of SP-1 and a subsequent increase in AS transcription. The regulati on of AL by glycosylation has not been studied. Taken together, the extensive stud ies of eNOS PTM by phosphorylation, nitrosylation, glycosylation and acylation have a prominent role in regulating the ability of the endothelium to produce NO. Due to the fe w papers in the lite rature that suggest that AS is also regulated by PTM and to the prominent role of serine/threonine phosphorylation in regulating eNOS functi on, Specific Aim 2 is dedicated to the thorough investigation of endot helial AS serine/threonine phosphorylation. The findings are presented in Chapter Two. Regulation of Endothelial Nitric Oxide Pr oduction: Subcellular Localization and Protein Interactions The precise location of a protein within the cell is critical for its function. Compartmentalization of proteins that are in the same metabolic or signaling pathway allows for efficient communication, channe ling of substrates and accessibility to important regulatory factors. In addition, th e regulation of subce llular transport to localize these proteins invol ves several mechanisms including vesicles, cytoskeletal components and protein interactions. Furthermor e, protein interacti ons often mediate the functions of their interacting partners via m echanisms that do not involve transport such as post-translational modification or conformational change [216]. There is a complex

PAGE 45

27 network of interacting proteins that are known to regulat e NO production via eNOS [180] and the scientific community is just beginning to de scribe similar mechanisms for AS and AL. The regulation of eNOS is dependent its localization with in the cell [217, 218]. There are two regions in the cell that have been found to independently produce NO, the Golgi and the plasma membrane [219, 220]. Within the plasma membrane, eNOS is localized to caveolae [217]. Ca veolae are invaginations of th e plasma membrane that are rich in glycosphingolipids and cholesterol and serve as platforms for the integration of signaling pathways. Caveolae are also involved in endocytosis and transcytosis. Caveolin is an integral membrane protein that is an important structural component of caveolae [218, 221]. Although activation of eNOS by phosphorylation at S1179 does not cause translocation from the Golgi to the plasma membrane, the proper localization of eNOS is necessary for this phosphorylation to o ccur [208, 220]. Mistargeting of eNOS also attenuates NO release [222, 223]. In the plasma membrane, eNOS is constitutively phosphorylated at S1179 and is highly active. In the Golgi, phosphorylation at S1179 is diminished and less NO is produced in respons e to calcium signaling [224]. Furthermore, plasma membrane eNOS is more responsive to agonists such as in sulin and angiopoietin and demonstrates increased binding to HSP90 as compared to Golgi-eNOS [219]. Despite this understanding, there is stil l some doubt as to whether Golgi versus plasma membrane NO production is of more or less significance or whether the distinct functions of these pools require further study to define their specific purpose in the cell.

PAGE 46

28 The targeting of eNOS to Golgi versus pl asma membrane is controlled by several mechanisms. One key mechanism involves myristoylation and palmitoylation [209]. Myristoylation is necessary for eNOS to associate with the membrane, essential for optimal eNOS activity, important for comp artmentalization in the Golgi and is a prerequisite for palmitoylation. Palmitoylati on is essential for e NOS targeting to the membrane and specifically to caveolae [210] In a study by Jagnandan and colleagues, organelle-specific eNOS constr ucts were generated in an attempt to determine the compartment-specific function of eNOS. The authors found that the nucleus, mitochondria cytoplasm and trans-Golgi ne twork are inefficient for NO production as compared to the plasma membrane and cis-Go lgi network. This is ap parently not due to insufficient substrate availability since iNOS targeted to the same regions functions normally. The relative lack of calcium/calmodulin in these regions may responsible for the inefficiency of eNOS, which unlike iNOS is dependent on calcium and calmodulin for function [224]. A subsequent study determ ined that calcium and calmodulin are not the reason for disparate function in different organelles. The study found it is actually phosphorylation that regulates organelle-speci fic eNOS function since Golgi localized eNOS had reduced phosphorylati on at S1179, S635 and S617 [225]. In addition to acylation as a targeting mechanism, there is an intricate network of cytoskeletal components that regulate eNOS function and like ly also its translocation. Actin is a prominent regulator of eNOS. For ex ample, F-actin is required for shear stress mediated eNOS upregulation due to its role in mechanotransduction [226]. In general, the interaction of eNOS with actin increases it s activity. One mechanism is due to increased

PAGE 47

29 interaction with HSP90 [227]. On the other hand, the inte raction of eNOS with actinin-4, an actin-associated protein, inhibits eNOS activity by preventing calmodulin from interacting with and activati ng eNOS [228]. Vimentin is another important cytoskeletal component that regulates eNOS function. Vimen tin is an intermediate filament that is essential for flow-mediated d ilatation [226]. Alt hough the eNOS activity per se has not been related to vimentin, th e importance of this cytoskel etal component in mediating vessel relaxation implies the involvement of eNOS. Vimentin is also an essential structural component of caveolae [229]. In addition, the inhibition of microtubules (tubulin and microtubule associ ated proteins) leads to decreased eNOS activity and NO production [230]. Furthermore, it has been de monstrated that HSP90 and calmodulin both interact with tubulin. The interaction of tubulin with HSP90 is important for eNOS activity since disruption of microtubule assembly diminishes the eNOS-HSP90 interaction and reduces NO output The specific role of the calmodulin-tubulin interaction on the function of the citrulline-NO cycle is unclear [230]. As discussed in a latter section, both HSP90 and calmodulin are impor tant binding partners of eNOS that stimulate its function. This suggests that microt ubules also play a role in eNOS function. Whether the mechanism is structural or involves eNOS transport is not clear. Finally, several protein interactions are essential for eNOS cellular transport. One well studied interaction is with caveolin-1. AS discussed previously, caveolin-1 is an important component of caveolae. It is believed that caveol ae function as vesicles that regulate eNOS transcytosis and caveolin-1 is required in this process [222]. More importantly, caveolin-1 and eNOS both in teract with NOSTRIN (eNOS trafficking

PAGE 48

30 inducer), which also recruits dynamin, a Golgi localized GTP-ase, a nd facilitates caveolar transport [231]. NOSTRIN transport of eNOS is dependent on the actin cytoskeleton [232]. Although NOSTRIN is important for eNOS translocation, it inhibits eNOS activity regardless of localization in the cell and overexpression of NOSTRIN leads to redistribution of eNOS from the plasma membra ne to intracellular vesicular structures [232]. The specific regulation of this complex and its trafficking is still unclear. NOSIP (eNOS interacting protein) is another important protein that regulates eNOS translocation. NOSIP interacts directly w ith eNOS and inhibits its activity [233]. Furthermore, NOSIP is important for the subc ellular redistribution of eNOS between the caveolae and intracellular compartmen ts [234]. Much like NOSTRIN, NOSIP overexpression leads diminished plasma memb rane localization of eNOS [233]. To date, there have been no studies de lineating the regulati on of AS or AL intracellular targeting. The lo calization of AS and AL has been understudied in endothelial cells. However, we have demonstrated that AS, AL eNOS and caveolin-1 cofractionate in caveolar membrane preparatio ns [110]. In vascular smooth muscle, AS demonstrates both cytoplasmic and membrane localization along with a punctuate pattern of distribution that suggests mitochondrial locali zation [108]. Otherwise, no other studies have addressed the subcellular localization of AS or AL as part of the citrulline-NO cycle. Much more is known about their localization in tis sues that produce urea. For example, in liver, these enzymes are localized in mitochondria.

PAGE 49

31 In addition to the role protein interactions play in the targeting of eNOS, there are several proteins that regulate eNOS functions in other wa ys. Aside from the role of caveolin in AS localization, it has direct e ffects on inhibiting eNOS activity [235]. The interaction of caveolin with eNOS is direct and occurs in a region of eNOS that contains a caveolin binding motif [235]. Th e interaction of eNOS with caveolin is disrupted upon stimulation with calcium i onophores, bradykinin or fluid sh ear stress [236, 237]. Another interaction that negatively re gulates eNOS function is th e bradykinin receptor. This interaction is similar to that with caveo lin in that it represents a membrane docking interaction that is relieved upon treatment w ith agonists such as bradykinin and calcium ionophores [238]. There are a multitude of proteins that interact with eNOS and enhance its function. One prominent interaction is w ith heat shock protein-90 (HSP90). This molecular chaperone activates eNOS via a mechanism that involves phosphorylation at S1179 [239]. HSP90 delivers the kinase Akt to eNOS and promotes its phosphorylation [240]. Calmodulin is an importa nt positive regulator of eNOS that also interacts with HSP90 and is responsible for displacing the eNOS-caveolin interaction [241]. In fact, caveolin and calmodulin are involved in the reciprocal regulation of eNOS. Upon agonist stimulation and an increase in intracellular calcium, calmodulin binds to eNOS and displaces caveolin [242]. In a ddition, eNOS interacts with the CAT-1 arginine transporter and this increases its activation via phos phorylation at S1179 and 635. The mechanism of enhanced NO release does not involve arginine transport [243].

PAGE 50

32 In contrast to eNOS, very little is know n about protein intera ctions involving AS and AL as part of the citrulline-NO cycle. Recently, the interaction of an NADPH sensor protein (HSCARG) with AS was shown to dow n regulate AS activity in epithelial cells. Their results implied that HSCARG regulation of AS activity is crucial for maintaining the intracellular balan ce between redox state and NO levels [244]. This is the first study to define a protein interaction with AS that is essential for NO production. Given the prominence of regulation of NO production by eNOS subcellular localization and protein intera ctions, we hypothesize similar mechanisms exist to regulate the role of AS in endothelial NO producti on. Therefore, Specific Aim 3 of this dissertation will explore both the subcellular lo calization and protein interactions of AS. Our findings are described in Chapter Three. References [1] A. J. Gow, and H. Ischiropoulos, Nitric oxide chemistry and cellular signaling, J Cell Physiol 187 (2001) 277-282. [2] J. S. Stamler, D. J. Singel, and J. Los calzo, Biochemistry of nitric oxide and its redoxactivated forms, Science 258 (1992) 1898-1902. [3] E. Culotta, and D. E. Koshland, Jr., NO news is good news, Science 258 (1992) 18621865. [4] D. E. Koshland, Jr., The molecule of the year, Science 258 (1992) 1861. [5] R. F. Furchgott, and J. V. Zawadzki, The obligatory role of endothelial cells in the relaxation of arterial smooth muscle by acetylcholine, Nature 288 (1980) 373-376.

PAGE 51

33 [6] L. J. Ignarro, G. M. Buga, K. S. Wood, R. E. Byrns, and G. Chaudhuri, Endotheliumderived relaxing factor produced and released from artery and vein is nitric oxide, Proc Natl Acad Sci U S A 84 (1987) 9265-9269. [7] R. M. Palmer, A. G. Ferrige, and S. Moncada, Nitric oxide re lease accounts for the biological activity of endothelium-derived relaxing factor, Nature 327 (1987) 524-526. [8] S. Moncada, Nitric oxide in the vasc ulature: physiology a nd pathophysiology, Ann N Y Acad Sci 811 (1997) 60-67; discussion 67-69. [9] S. Moncada, and E. A. Higgs, The discovery of nitric oxide and its role in vascular biology, Br J Pharmacol 147 Suppl 1 (2006) S193-201. [10] W. S. Jobgen, S. K. Fried, W. J. Fu, C. J. Meininger, and G. Wu, Regulatory role for the arginine-nitric oxide pathway in metabolism of energy substrates, J Nutr Biochem 17 (2006) 571-588. [11] G. Wu, and S. M. Mo rris, Jr., Arginine metabolis m: nitric oxide and beyond, Biochem J 336 ( Pt 1) (1998) 1-17. [12] W. E. O'Brien, Isolati on and characterization of argininosuccinate synthetase from human liver, Biochemistry 18 (1979) 5353-5356. [13] S. Ratner, Enzymes of arginine and urea synthesis, Adv Enzymol Relat Areas Mol Biol 39 (1973) 1-90. [14] T. S. Su, H. G. Bock, W. E. O'Brie n, and A. L. Beaudet, Cloning of cDNA for argininosuccinate synthetase mRNA and study of enzyme overproduction in a human cell line, J Biol Chem 256 (1981) 11826-11831. [15] O. Rochovansky, H. Kodowaki, and S. Ra tner, Biosynthesis of urea. Molecular and regulatory properties of crys talline argininosuccinate synthetase, J Biol Chem 252 (1977) 5287-5294. [16] O. Rochovansky, and S. Ratner, Biosynt hesis of urea. XII. Further studies on argininosuccinate synthetase: substrate affinity and mechan ism of action, J Biol Chem 242 (1967) 3839-3849. [17] M. J. Jackson, A. L. Beaudet, and W. E. O'Brien, Mammalian urea cycle enzymes, Annu Rev Genet 20 (1986) 431-464. [18] W. E. O'Brien, R. McInnes, K. Kalu muck, and M. Adcock, Cloning and sequence analysis of cDNA for human argininosuccina te lyase, Proc Natl Acad Sci U S A 83 (1986) 7211-7215.

PAGE 52

34 [19] W. K. Alderton, C. E. Coope r, and R. G. Knowles, Nitric oxide synthases: structure, function and inhibition, Biochem J 357 (2001) 593-615. [20] T. O. Fischmann, A. Hruza, X. D. Niu, J. D. Fossetta, C. A. Lunn, E. Dolphin, A. J. Prongay, P. Reichert, D. J. Lundell, S. K. Narula, and P. C. Weber, Structural characterization of nitric oxide synthase isoforms re veals striking active-site conservation, Nat Struct Biol 6 (1999) 233-242. [21] D. S. Bredt, and S. H. Snyder, Isolat ion of nitric oxide s ynthetase, a calmodulinrequiring enzyme, Proc Natl Acad Sci U S A 87 (1990) 682-685. [22] S. Lamas, P. A. Marsden, G. K. Li, P. Tempst, and T. Michel, Endothelial nitric oxide synthase: molecular clon ing and characterization of a distinct constitutive enzyme isoform, Proc Natl Acad Sci U S A 89 (1992) 6348-6352. [23] K. K. McDonald, S. Zhar ikov, E. R. Block, and M. S. Kilberg, A caveolar complex between the cationic amino acid transporter 1 and endothelial nitric-oxide synthase may explain the "arginine paradox", J Biol Chem 272 (1997) 31213-31216. [24] B. R. Flam, D. C. Eichler, and L. P. Solomonson, Endothelial nitric oxide production is tightly coupled to the citrul line-NO cycle, Nitric Oxide 17 (2007) 115-121. [25] C. W. Shuttleworth, A. J. Burns, S. M. Ward, W. E. O'Brien, and K. M. Sanders, Recycling of L-citrulline to sustain nitric oxide-dependent enteric neurotransmission, Neuroscience 68 (1995) 1295-1304. [26] L. J. Shen, K. Beloussow, and W. C. Shen, Accessibility of endothelial and inducible nitric oxide synthase to the intracellula r citrulline-arginine regeneration pathway, Biochem Pharmacol 69 (2005) 97-104. [27] G. Wu, and C. J. Meini nger, Regulation of L-arginine synthesis from L-citrulline by L-glutamine in endothelial cells, Am J Physiol 265 (1993) H1965-1971. [28] M. Hecker, W. C. Sessa, H. J. Harris, E. E. Anggard, and J. R. Vane, The metabolism of L-arginine and its significance for the biosynthesis of endothelium-derived relaxing factor: cultured endothe lial cells recycle L-citrulline to L-arginine, Proc Natl Acad Sci U S A 87 (1990) 8612-8616. [29] J. L. Deignan, S. D. Cederbaum, and W. W. Grody, Contrasting features of urea cycle disorders in human patients and knockout mouse models, Mol Genet Metab 93 (2008) 7-14. [30] J. A. Dennis, P. J. Healy, A. L. Beaude t, and W. E. O'Brien, Molecular definition of bovine argininosuccinate s ynthetase deficiency, Proc Natl Acad Sci U S A 86 (1989) 7947-7951.

PAGE 53

35 [31] J. Haberle, S. Pauli, M. Linnebank, W. J. Kleijer, H. D. Bakker, R. J. Wanders, E. Harms, and H. G. Koch, Structure of the human argininosuccinate synthetase gene and an improved system for molecular diagnostics in patients with classical and mild citrullinemia, Hum Genet 110 (2002) 327-333. [32] K. Kobayashi, M. J. Jackson, D. B. Tick, W. E. O'Brien, and A. L. Beaudet, Heterogeneity of mutations in argininosuccina te synthetase causing human citrullinemia, J Biol Chem 265 (1990) 11361-11367. [33] K. Kobayashi, D. S. Sinasac, M. Iijima A. P. Boright, L. Begum, J. R. Lee, T. Yasuda, S. Ikeda, R. Hirano, H. Terazono, M. A. Crackower, I. Kondo, L. C. Tsui, S. W. Scherer, and T. Saheki, The ge ne mutated in adult-onset type II citrullinaemia encodes a putative mitochondrial carrier protein, Nat Genet 22 (1999) 159-163. [34] T. Saheki, and K. Kobayashi, Mitochond rial aspartate glutam ate carrier (citrin) deficiency as the cause of adult-onset type II citrullinemia (CTLN2) and idiopathic neonatal hepatitis (NICCD), J Hum Genet 47 (2002) 333-341. [35] G. M. Enns, S. A. Berry, G. T. Berry, W. J. Rhead, S. W. Brusilow, and A. Hamosh, Survival after treatment with phenylacetate and benzoate for urea-cycle disorders, N Engl J Med 356 (2007) 2282-2292. [36] V. Reid Sutton, Y. Pan, E. C. Davi s, and W. J. Craigen, A mouse model of argininosuccinic aciduria: biochemical characterization, Mol Genet Metab 78 (2003) 1116. [37] M. Sausbier, R. Schubert, V. Voigt, C. Hirneiss, A. Pfeifer, M. Korth, T. Kleppisch, P. Ruth, and F. Hofmann, Mechanisms of NO/cGMP-dependent vasorelaxation, Circ Res 87 (2000) 825-830. [38] A. Pyriochou, and A. Papapetropoulos, Soluble guanylyl cyclase: more secrets revealed, Cell Signal 17 (2005) 407-413. [39] D. Alonso, and M. W. Radomski, Nitr ic oxide, platelet function, myocardial infarction and reperfusion th erapies, Heart Fail Rev 8 (2003) 47-54. [40] I. Fleming, C. Schulz, B. Fichtlscherer, B. E. Kemp, B. Fisslthaler, and R. Busse, AMP-activated protein kinase (AMPK) regul ates the insulin-induced activation of the nitric oxide synthase in huma n platelets, Thromb Haemost 90 (2003) 863-871. [41] J. P. Cooke, NO and angi ogenesis, Atheroscler Suppl 4 (2003) 53-60. [42] M. Ziche, and L. Morbidelli, Nitr ic oxide and angiogene sis, J Neurooncol 50 (2000) 139-148.

PAGE 54

36 [43] M. Cudmore, S. Ahmad, B. Al-Ani, P. Hewett, S. Ahmed, and A. Ahmed, VEGF-E activates endothelial nitric oxide synthase to induce a ngiogenesis via cGMP and PKGindependent pathways, Bi ochem Biophys Res Commun 345 (2006) 1275-1282. [44] A. Parenti, L. Mo rbidelli, F. Ledda, H. J. Granger, and M. Ziche, The bradykinin/B1 receptor promotes angiogenesis by up-regulation of endogenous FGF-2 in endothelium via the nitric oxide synthase pathway, Faseb J 15 (2001) 1487-1489. [45] F. N. Kiefer, S. Neysari, R. Humar, W. Li, V. C. Munk, and E. J. Battegay, Hypertension and angiogenesis, Curr Pharm Des 9 (2003) 1733-1744. [46] M. Kuwabara, Y. Kakinuma, M. Ando, R. G. Katare, F. Yamasaki, Y. Doi, and T. Sato, Nitric oxide stimulates vascular endothelial growth factor production in cardiomyocytes involved in angiogenesis, J Physiol Sci 56 (2006) 95-101. [47] W. Zheng, E. A. Seftor, C. J. Mein inger, M. J. Hendrix, and R. J. Tomanek, Mechanisms of coronary angiogenesis in res ponse to stretch: role of VEGF and TGFbeta, Am J Physiol Heart Circ Physiol 280 (2001) H909-917. [48] A. Papapetropoulos, G. Ga rcia-Cardena, J. A. Madri, an d W. C. Sessa, Nitric oxide production contributes to the angiogenic properties of vascular endothelial growth factor in human endothelial ce lls, J Clin Invest 100 (1997) 3131-3139. [49] J. X. Chen, M. L. Lawrence, G. C unningham, B. W. Christman, and B. Meyrick, HSP90 and Akt modulate Ang-1-induced angi ogenesis via NO in coronary artery endothelium, J Appl Physiol 96 (2004) 612-620. [50] M. Ruel, R. S. Beanlands, M. Lortie, V. Chan, N. Camack, R. A. deKemp, E. J. Suuronen, F. D. Rubens, J. N. DaSilva, F. W. Sellke, D. J. Stewart, and T. G. Mesana, Concomitant treatment with oral L-argini ne improves the efficacy of surgical angiogenesis in patients with severe diffuse coronary arte ry disease: the Endothelial Modulation in Angiogenic Therapy randomized controlled trial, J Thorac Cardiovasc Surg 135 (2008) 762-770, 770 e761. [51] G. Melino, M. V. Catani, M. Corazzari, P. Guerrieri, and F. Bernassola, Nitric oxide can inhibit apoptosis or switch it into necrosis, Cell Mol Life Sci 57 (2000) 612-622. [52] S. Dimmeler, and A. M. Zeiher, Nitric oxide and apoptosis: another paradigm for the double-edged role of nitric oxide, Nitric Oxide 1 (1997) 275-281. [53] Y. H. Shen, X. L. Wang, and D. E. Wilcken, Nitric oxide induces and inhibits apoptosis through different pathways, FEBS Letters 433 (1998) 125-131.

PAGE 55

37 [54] H. T. Chung, H. O. Pae, B. M. Choi, T. R. Billiar, and Y. M. Kim, Nitric oxide as a bioregulator of apoptosis, Biochem Biophys Res Commun 282 (2001) 1075-1079. [55] N. Makino, T. Maeda, M. Sugano, S. Sa toh, R. Watanabe, and N. Abe, High serum TNF-alpha level in Type 2 diabetic patient s with microangiopathy is associated with eNOS down-regulation and apoptosis in endothelial cells, J Diabetes Complications 19 (2005) 347-355. [56] N. Andrieu-Abadie, V. Gouaze, R. Salv ayre, and T. Levade, Ceramide in apoptosis signaling: relationship with oxidativ e stress, Free Ra dic Biol Med 31 (2001) 717-728. [57] C. J. Gamard, G. S. Dbaibo, B. Liu, L. M. Obeid, and Y. A. Hannun, Selective involvement of ceramide in cytokine-induced apoptosis. Ceramide inhibits phorbol ester activation of nuclear factor kappaB, J Biol Chem 272 (1997) 16474-16481. [58] Y. Ido, D. Carling, and N. Ruderman, Hyperglycemia-induced apoptosis in human umbilical vein endothelial cells: inhibition by the AMP-activated protein kinase activation, Diabetes 51 (2002) 159-167. [59] C. Blazquez, M. J. Geelen, G. Velasc o, and M. Guzman, The AMP-activated protein kinase prevents ceramide synthesis de novo and apoptosis in astrocytes, FEBS Lett 489 (2001) 149-153. [60] T. Matsunaga, S. Kotamraju, S. V. Kalivendi, A. Dhanasekaran, J. Joseph, and B. Kalyanaraman, Ceramide-induced intracellu lar oxidant formation, iron signaling, and apoptosis in endothelial cells: protective ro le of endogenous nitric oxide, J Biol Chem 279 (2004) 28614-28624. [61] S. Dimmeler, J. Haendeler, V. Rippmann, M. Nehls, and A. M. Zeiher, Shear stress inhibits apoptosis of human e ndothelial cells, FEBS Letters 399 (1996) 71-74. [62] S. Dimmeler, C. Hermann, J. Galle, and A. M. Zeiher, Upregulation of superoxide dismutase and nitric oxide synthase mediates the apoptosis-suppressive effects of shear stress on endothelial cells, Arte rioscler Thromb Vasc Biol 19 (1999) 656-664. [63] U. Forstermann, and T. Munzel, Endothe lial nitric oxide s ynthase in vascular disease: from marvel to menace, Circulation 113 (2006) 1708-1714. [64] U. Landmesser, B. Hornig, and H. Dr exler, Endothelial f unction: a critical determinant in atherosc lerosis?, Circulation 109 (2004) II27-33. [65] C. Napoli, F. de Nigris, S. Williams-Ignarro, O. Pignalosa, V. Sica, and L. J. Ignarro, Nitric oxide and atherosclerosi s: an update, Nitric Oxide 15 (2006) 265-279.

PAGE 56

38 [66] R. Schulz, T. Rassaf, P. B. Massion, M. Kelm, and J. L. Balligand, Recent advances in the understanding of the role of nitric oxi de in cardiovascular homeostasis, Pharmacol Ther 108 (2005) 225-256. [67] M. E. Widlansky, N. Gokce, J. F. Keaney, Jr., and J. A. Vita, The clinical implications of endothelial dysfunction, J Am Coll Cardiol 42 (2003) 1149-1160. [68] R. Ross, Atherosclerosis is an inflammatory disease, Am Heart J 138 (1999) S419420. [69] B. R. Clapp, G. M. Hirschfield, C. Storr y, J. R. Gallimore, R. P. Stidwill, M. Singer, J. E. Deanfield, R. J. MacAllister, M. B. Pepys, P. Vallance, and A. D. Hingorani, Inflammation and endothelial f unction: direct vascular e ffects of human C-reactive protein on nitric oxide bioavailability, Circulation 111 (2005) 1530-1536. [70] F. Kim, B. Gallis, and M. A. Corson, TN F-alpha inhibits flow and insulin signaling leading to NO production in aortic endotheli al cells, Am J Physiol Cell Physiol 280 (2001) C1057-1065. [71] G. Torre-Amione, S. Kapadia, J. Lee, J. B. Durand, R. D. Bies, J. B. Young, and D. L. Mann, Tumor necrosis factor-alpha and tumo r necrosis factor receptors in the failing human heart, Circulation 93 (1996) 704-711. [72] G. S. Hotamisligil, D. L. Murray, L. N. Choy, and B. M. Spiegelman, Tumor necrosis factor alpha inhibits signaling from the insulin receptor, Proc Natl Acad Sci U S A 91 (1994) 4854-4858. [73] G. Winkler, P. Lakatos, F. Salamon, Z. Nagy, G. Speer, M. Kovacs, G. Harmos, O. Dworak, and K. Cseh, Elevated serum TNF-al pha level as a link between endothelial dysfunction and insulin resistance in normo tensive obese patients, Diabet Med 16 (1999) 207-211. [74] M. Fujita, R. J. Mason, C. Cool, J. M. Shannon, N. Hara, and K. A. Fagan, Pulmonary hypertension in TNF-alpha-overexpre ssing mice is associated with decreased VEGF gene expression, J Appl Physiol 93 (2002) 2162-2170. [75] L. Agnoletti, S. Curello, T. Bachet ti, F. Malacarne, G. Gaia, L. Comini, M. Volterrani, P. Bonetti, G. Parrinello, M. Cadei, P. G. Grigolato, and R. Ferrari, Serum from patients with severe heart failure downre gulates eNOS and is proapoptotic: role of tumor necrosis factor -alpha, Circulation 100 (1999) 1983-1991. [76] F. M. Martens, T. J. Rabelink, J. Op 't Roodt, E. J. de Koning, and F. L. Visseren, TNF-{alpha} induces endothelial dysfunction in diabetic adults, an effect reversible by the PPAR-{gamma} agonist pioglitazone, Eur Heart J 27 (2006) 1605-1609.

PAGE 57

39 [77] K. Kajita, T. Mune, Y. Kanoh, Y. Natsume, M. Ishizawa, Y. Kawai, K. Yasuda, C. Sugiyama, and T. Ishizuka, TNFalpha reduces the expression of per oxisome proliferatoractivated receptor gamma ( PPARgamma) via the production of ceramide and activation of atypical PKC, Diabetes Res Clin Pract 66 Suppl 1 (2004) S79-83. [78] C. Chen, V. A. Korshunov, M. P. Masse tt, C. Yan, and B. C. Berk, Impaired vasorelaxation in inbred mice is associated w ith alterations in both nitric oxide and super oxide pathways, J Vasc Res 44 (2007) 504-512. [79] M. I. Lin, D. Fulton, R. Babbitt, I. Flem ing, R. Busse, K. A. Pritchard, Jr., and W. C. Sessa, Phosphorylation of threoni ne 497 in endothelial nitric-oxide synthase coordinates the coupling of L-arginine metabolism to efficient nitric oxide production, J Biol Chem 278 (2003) 44719-44726. [80] T. Munzel, A. Daiber, V. Ullrich, a nd A. Mulsch, Vascular consequences of endothelial nitric oxide syntha se uncoupling for the activity and expression of the soluble guanylyl cyclase and the cGMP -dependent protein kinase, Arterioscler Thromb Vasc Biol 25 (2005) 1551-1557. [81] C. A. Chen, L. J. Druhan, S. Varadharaj, Y. R. Chen, and J. L. Zweier, Phosphorylation of endothelial nitric-oxide synthase regul ates superoxide generation from the enzyme, J Biol Chem 283 (2008) 27038-27047. [82] A. J. Cardounel, H. Cui, A. Samouilov, W. Johnson, P. Kearns, A. L. Tsai, V. Berka, and J. L. Zweier, Evidence for the pathophysiological role of endogenous methylarginines in regulation of endothelia l NO production and vascular function, J Biol Chem 282 (2007) 879-887. [83] N. Sud, S. M. Wells, S. Sharma, D. A. Wiseman, J. Wilham, and S. M. Black, Asymmetric dimethylarginine inhibits HSP90 activity in pu lmonary arterial endothelial cells: role of mitochondr ial dysfunction, Am J Physiol Cell Physiol 294 (2008) C14071418. [84] M. C. Stuhlinger, E. Conci, B. J. Ha ubner, E. M. Stocker, J. Schwaighofer, J. P. Cooke, P. S. Tsao, O. Pachinger, and B. Metzler, Asymmetric dimethyl L-arginine (ADMA) is a critical regulator of myocar dial reperfusion injury, Cardiovasc Res 75 (2007) 417-425. [85] H. M. Eid, T. Lyberg, H. Arnesen, and I. Seljeflot, Insulin and adiponectin inhibit the TNFalpha-induced ADMA accumulation in hu man endothelial cells: the role of DDAH, Atherosclerosis 194 (2007) e1-8. [86] L. J. Druhan, S. P. Forbes, A. J. Pope, C. A. Chen, J. L. Zweier, and A. J. Cardounel, Regulation of eNOS-derived superoxide by endogenous methylarginines, Biochemistry 47 (2008) 7256-7263.

PAGE 58

40 [87] V. De Gennaro Colonna, S. Bonomo, P. Ferrario, M. Bianchi, M. Berti, M. Guazzi, B. Manfredi, E. E. Muller, F. Berti, a nd G. Rossoni, Asymmetric dimethylarginine (ADMA) induces vascular endothelium im pairment and aggravates post-ischemic ventricular dysfunction in rats, Eur J Pharmacol 557 (2007) 178-185. [88] J. Zsuga, J. Torok, M. T. Magyar, A. Va likovics, R. Gesztelyi, S. Keki, L. Csiba, M. Zsuga, and D. Bereczki, Serum asymmetric di methylarginine negatively correlates with intima-media thickness in early-onset atherosclerosis, Cerebrovasc Dis 23 (2007) 388394. [89] A. Blair, P. W. Shaul, I. S. Yuhanna, P. A. Conrad, and E. J. Smart, Oxidized low density lipoprotein displaces endothelia l nitric-oxide synthase (eNOS) from plasmalemmal caveolae and impairs eNOS activation, J Biol Chem 274 (1999) 3251232519. [90] H. Li, P. Junk, A. Huwiler, C. Burkhard t, T. Wallerath, J. Pfeilschifter, and U. Forstermann, Dual effect of ceramide on huma n endothelial cells: induction of oxidative stress and transcriptional upr egulation of endothelial nitric oxide synthase, Circulation 106 (2002) 2250-2256. [91] E. Clementi, N. Borgese, and J. Meldol esi, Interactions between nitric oxide and sphingolipids and the potential consequences in physiology and pathology, Trends Pharmacol Sci 24 (2003) 518-523. [92] A. Kawakami, M. Osaka, M. Tani, H. Azuma, F. M. Sacks, K. Shimokado, and M. Yoshida, Apolipoprotein CIII links hyperlip idemia with vascular endothelial cell dysfunction, Circulation 118 (2008) 731-742. [93] T. Fujii, M. Onimaru, Y. Yonemitsu, H. Kuwano, and K. Sueishi, Statins restore ischemic limb blood flow in diabetic micr oangiopathy via eNOS/NO upregulation but not via PDGF-BB expression, Am J Ph ysiol Heart Circ Physiol 294 (2008) H2785-2791. [94] G. I. Boger, T. K. Rudolph, R. Maas, E. Schwedhelm, E. Dumbadze, A. Bierend, R. A. Benndorf, and R. H. Boger, Asymmetr ic dimethylarginine determines the improvement of endothelium-dependent vasodilation by simvastatin: Effect of combination with oral L-arginine, J Am Coll Cardiol 49 (2007) 2274-2282. [95] J. A. Kim, M. Montagnani, K. K. K oh, and M. J. Quon, Reciprocal relationships between insulin resistance and endothelial dys function: molecular and pathophysiological mechanisms, Circulation 113 (2006) 1888-1904. [96] S. E. Borst, The ro le of TNF-alpha in insuli n resistance, Endocrine 23 (2004) 177182.

PAGE 59

41 [97] F. Bourgoin, H. Bachelard, M. Badeau, S. Melancon, M. Pitre, R. Lariviere, and A. Nadeau, Endothelial and vascular dysfunctions and insulin resistance in rats fed a highfat, high-sucrose diet, Am J Physiol Heart Circ Physiol 295 (2008) H1044-H1055. [98] G. Doronzo, I. Russo, L. Mattiello, G. Anfossi, A. Bosia, and M. Trovati, Insulin activates vascular endothelial gr owth factor in vascular smoot h muscle cells: influence of nitric oxide and of insulin re sistance, Eur J Clin Invest 34 (2004) 664-673. [99] R. D. Feldman, and G. S. Bierbrier, In sulin-mediated vasodilation: impairment with increased blood pressure and body mass, Lancet 342 (1993) 707-709. [100] S. Stratford, K. L. Hoehn, F. Liu, and S. A. Summers, Regulation of insulin action by ceramide: dual mechanisms linking ceramide accumulation to the inhibition of Akt/protein kinase B, J Biol Chem 279 (2004) 36608-36615. [101] X. L. Du, D. Edelstein, S. Dimmele r, Q. Ju, C. Sui, and M. Brownlee, Hyperglycemia inhibits endothelial nitric oxi de synthase activity by posttranslational modification at the Akt site, J Clin Invest 108 (2001) 1341-1348. [102] Y. Ding, N. D. Vaziri, R. Coulson, V. S. Kamanna, and D. D. Roh, Effects of simulated hyperglycemia, insulin, and glu cagon on endothelial nitric oxide synthase expression, Am J Physiol Endocrinol Metab 279 (2000) E11-17. [103] L. H. Pojoga, T. M. Yao, S. Sinha, R. L. Ross, J. C. Lin, J. D. Raffetto, G. K. Adler, G. H. Williams, and R. A. Khalil, Eff ect of dietary sodium on vasoconstriction and eNOS-mediated vascular relaxation in caveo lin-1-deficient mice, Am J Physiol Heart Circ Physiol 294 (2008) H1258-1265. [104] R. De Caterina, A. Zampolli, S. Del Turco, R. Madonna, and M. Massaro, Nutritional mechanisms that influence car diovascular disease, Am J Clin Nutr 83 (2006) 421S-426S. [105] W. Z. Zhang, K. Venardos, J. Chin-D usting, and D. M. Kaye, Adverse effects of cigarette smoke on NO bioavailab ility: role of arginine me tabolism and oxidative stress, Hypertension 48 (2006) 278-285. [106] R. P. Mason, Scientific rationale for combination of a calcium channel antagonist and an HMG-CoA reductase inhi bitor: a new approach to risk factor management, Drugs 68 (2008) 885-900. [107] L. Xie, Y. Hattori, N. Tume, and S. S. Gross, The preferred source of arginine for high-output nitric oxide synthesis in blood vessels, Semin Perinatol 24 (2000) 42-45.

PAGE 60

42 [108] L. Xie, and S. S. Gross, Argininosucci nate synthetase overe xpression in vascular smooth muscle cells poten tiates immunostimulant-induced NO production, J Biol Chem 272 (1997) 16624-16630. [109] B. L. Goodwin, L. P. Solomonson, and D. C. Eichler, Argininosuccinate synthase expression is required to maintain nitric ox ide production and cell viability in aortic endothelial cells, J Biol Chem 279 (2004) 18353-18360. [110] B. R. Flam, P. J. Hartmann, M. Ha rrell-Booth, L. P. Solomonson, and D. C. Eichler, Caveolar localizati on of arginine regeneration enzymes, argininosuccinate synthase, and lyase, with endothelial nitric oxide synthase, Nitric Oxide 5 (2001) 187197. [111] L. C. Pendleton, B. L. Goodwin, L. P. Solomonson, and D. C. Eichler, Regulation of endothelial argininosuccinate synthase expression and NO production by an upstream open reading frame, J Biol Chem 280 (2005) 24252-24260. [112] G. Hao, L. Xie, and S. S. Gross, Ar gininosuccinate synthetase is reversibly inactivated by S-nitrosylation in vi tro and in vivo, J Biol Chem 279 (2004) 36192-36200. [113] T. Karlberg, R. Collins, S. van de n Berg, A. Flores, M. Hammarstrom, M. Hogbom, L. Holmberg Schiavone, and J. Uppenberg, Structure of human argininosuccinate synthetase, Acta Crystallogr D Biol Crystallogr 64 (2008) 279-286. [114] R. Govers, and T. J. Rabelink, Cellular regulation of endothelial nitric oxide synthase, Am J Physiol Renal Physiol 280 (2001) F193-206. [115] D. S. Bredt, and S. H. Snyder, N itric oxide: a physiologic messenger molecule, Annu Rev Biochem 63 (1994) 175-195. [116] I. N. Mungrue, M. Husain, and D. J. St ewart, The role of NOS in heart failure: lessons from murine genetic models, Heart Fail Rev 7 (2002) 407-422. [117] H. Duplain, R. Burcelin, C. Sartori, S. Cook, M. Egli, M. Lepori, P. Vollenweider, T. Pedrazzini, P. Nicod, B. Thorens, and U. Scherrer, Insulin resistance, hyperlipidemia, and hypertension in mice lacking endothelial nitric oxide synthase, Circulation 104 (2001) 342-345. [118] I. Momken, D. Fortin, B. Serrurier, X. Bigard, R. Ventura-Clap ier, and V. Veksler, Endothelial nitric oxide synt hase (NOS) deficiency affect s energy metabolism pattern in murine oxidative skeletal muscle, Biochem J 368 (2002) 341-347.

PAGE 61

43 [119] R. van Haperen, M. de Waard, E. van Deel, B. Mees, M. Kutryk, T. van Aken, J. Hamming, F. Grosveld, D. J. Duncker, and R. de Crom, Reduction of blood pressure, plasma cholesterol, and athero sclerosis by elevated endothelial nitric oxide, J Biol Chem 277 (2002) 48803-48807. [120] J. W. Elrod, J. J. Greer, N. S. Bryan, W. Langston, J. F. Szot, H. Gebregzlabher, S. Janssens, M. Feelisch, and D. J. Lefer, Cardiomyocyte-specific overexpression of NO synthase-3 protects against myocardial isch emia-reperfusion injury, Arterioscler Thromb Vasc Biol 26 (2006) 1517-1523. [121] W. Shi, X. Wang, D. M. Shih, V. E. Laubach, M. Navab, and A. J. Lusis, Paradoxical reduction of fatty streak formati on in mice lacking endothelial nitric oxide synthase, Circulation 105 (2002) 2078-2082. [122] M. A. Talukder, T. Fujiki, K. Morikawa M. Motoishi, H. Kubota, T. Morishita, M. Tsutsui, A. Takeshita, and H. Shimokawa, Up-regulated neuronal nitric oxide synthase compensates coronary flow response to bradykinin in endothelial nitric oxide synthasedeficient mice, J Cardiovasc Pharmacol 44 (2004) 437-445. [123] F. Brunner, P. Andrew, G. Wolkart, R. Zechner, and B. Mayer, Myocardial contractile function and heart rate in mi ce with myocyte-speci fic overexpression of endothelial nitric oxide synthase, Circulation 104 (2001) 3097-3102. [124] M. Tsutsui, H. Shimokawa, T. Mori shita, Y. Nakashima, and N. Yanagihara, Development of genetically engineered mice l acking all three nitric oxide synthases, J Pharmacol Sci 102 (2006) 147-154. [125] X. Ye, B. Whiteman, M. Jerebtsova and M. L. Batshaw, Correction of argininosuccinate synthetase (A S) deficiency in a murine model of citrullinemia with recombinant adenovirus carrying human AS cDNA, Gene Ther 7 (2000) 1777-1782. [126] O. Traub, and B. C. Berk, Laminar sh ear stress: mechanisms by which endothelial cells transduce an atheroprotective for ce, Arterioscler Thromb Vasc Biol 18 (1998) 677685. [127] I. Fleming, and R. Busse, Molecular m echanisms involved in the regulation of the endothelial nitric oxide s ynthase, Am J Physiol Regul Integr Comp Physiol 284 (2003) R1-12. [128] C. D. Searles, Transc riptional and posttra nscriptional regulation of endothelial nitric oxide synthase expressi on, Am J Physiol Cell Physiol 291 (2006) C803-816. [129] P. L. Walpola, A. I. Gotlieb, M. I. Cybulsky, and B. L. Langille, Expression of ICAM-1 and VCAM-1 and monocyte adherence in arteries exposed to altered shear stress, Arterioscler Thromb Vasc Biol 15 (1995) 2-10.

PAGE 62

44 [130] B. Fisslthaler, S. Dimmeler, C. Hermann, R. Busse, and I. Fleming, Phosphorylation and activation of the endotheli al nitric oxide synt hase by fluid shear stress, Acta Physiologica Scandinavica 168 (2000) 81-88. [131] M. E. Davis, I. M. Grumbach, T. F ukai, A. Cutchins, and D. G. Harrison, Shear stress regulates endothelial nitric-oxide synt hase promoter activity through nuclear factor kappaB binding, J Biol Chem 279 (2004) 163-168. [132] M. E. Davis, H. Cai, G. R. Drumm ond, and D. G. Harrison, Shear stress regulates endothelial nitric oxide s ynthase expression through cSrc by divergent signaling pathways, Circ Res 89 (2001) 1073-1080. [133] M. Weber, C. H. Hagedorn, D. G. Harris on, and C. D. Searles, Laminar shear stress and 3' polyadenylation of eNOS mRNA, Circ Res 96 (2005) 1161-1168. [134] S. M. McCormick, S. G. Eskin, L. V. McIntire, C. L. Teng, C. M. Lu, C. G. Russell, and K. K. Chittur, DNA microarray re veals changes in gene expression of shear stressed human umbilical vein endothel ial cells, Proc Natl Acad Sci USA 98 (2001) 8955-8960. [135] B. P. Chan, W. M. Reichert, and G. A. Truskey, Synergistic effect of shear stress and streptavidin-biotin on the expression of endothelial vasodilator and cytoskeleton genes, Biotechnol Bioeng 88 (2004) 750-758. [136] D. J. MacEwan, TNF ligands and recep tors a matter of life and death, Br J Pharmacol 135 (2002) 855-875. [137] G. S. Hotamisligil, The role of TNFalpha and TNF receptors in obesity and insulin resistance, J Intern Med 245 (1999) 621-625. [138] S. H. Torres, J. B. De Sanctis, L. B. M. de, N. Hernandez, and H. J. Finol, Inflammation and nitric oxide pr oduction in skeletal muscle of type 2 diabetic patients, J Endocrinol 181 (2004) 419-427. [139] A. Aljada, H. Ghanim, E. Assian, and P. Dandona, Tumor necrosis factor-alpha inhibits insulin-induced increase in endothelial nitric oxide synthase and reduces insulin receptor content and phosphorylation in hum an aortic endothelial cells, Metabolism 51 (2002) 487-491. [140] T. de Frutos, L. S. de Miguel, M. Garcia-Duran, F. Gonzalez-Fernandez, J. A. Rodriguez-Feo, M. Monton, J. Guerra, J. Farre, S. Casado, and A. Lopez-Farre, NO from smooth muscle cells decreases NOS expression in endothelial cells: role of TNF-alpha, Am J Physiol 277 (1999) H1317-1325.

PAGE 63

45 [141] P. F. Lai, F. Mohamed, J. C. Monge, and D. J. Stewart, Downregulation of eNOS mRNA expression by TNFalpha: identificati on and functional characterization of RNAprotein interactions in the 3'UTR, Cardiovasc Res 59 (2003) 160-168. [142] M. Yoshizumi, M. A. Perrella, J. C. Burnett, Jr., and M. E. Lee, Tumor necrosis factor downregulates an endot helial nitric oxide synthase mRNA by shortening its halflife, Circ Res 73 (1993) 205-209. [143] H. D. Anderson, D. Rahmutula, and D. G. Gardner, Tumor necrosis factor-alpha inhibits endothelial nitric-oxide synthase gene promoter activity in bovine aortic endothelial cells, J Biol Chem 279 (2004) 963-969. [144] B. L. Goodwin, L. C. Pe ndleton, M. M. Levy, L. P. Solomonson, and D. C. Eichler, Tumor necrosis factor-alpha reduces argininosuccinate synthase expression and nitric oxide production in aortic endothelial cel ls, Am J Physiol Heart Circ Physiol 293 (2007) H1115-1121. [145] A. K. Nussler, T. R. Billia r, Z. Z. Liu, and S. M. Morris, Jr., Coinduction of nitric oxide synthase and argininos uccinate synthetase in a murine macrophage cell line. Implications for regulation of nitric oxide production, Journal of Biological Chemistry 269 (1994) 1257-1261. [146] M. Flodstrom, A. Niemann, F. J. Bedoya S. M. Morris, Jr., and D. L. Eizirik, Expression of the citrulline-nitr ic oxide cycle in rodent and human pancreatic beta-cells: induction of argininosucci nate synthetase by cytokines, Endocrinology 136 (1995) 32003206. [147] Y. Hattori, E. B. Campbell, and S. S. Gross, Argininosuccinate synthetase mRNA and activity are induced by immunostimulants in vascular smooth muscle. Role in the regeneration or arginine for nitric oxide synthesis, J Biol Chem 269 (1994) 9405-9408. [148] E. D. Rosen, and B. M. Spiegelm an, PPARgamma : a nuclear regulator of metabolism, differentiation, and cell growth, J Biol Chem 276 (2001) 37731-37734. [149] S. A. Kliewer, S. S. Sundseth, S. A. Jone s, P. J. Brown, G. B. Wisely, C. S. Koble, P. Devchand, W. Wahli, T. M. Willson, J. M. Lenhard, and J. M. Lehmann, Fatty acids and eicosanoids regulate gene expression thr ough direct interactions with peroxisome proliferator-activated receptors alpha and gamma, Pr oc Natl Acad Sci U S A 94 (1997) 4318-4323. [150] K. Schoonjans, J. Peinado-Onsurbe, A. M. Lefebvre, R. A. Heyman, M. Briggs, S. Deeb, B. Staels, and J. Auwerx, PPARalpha and PPARgamma activator s direct a distinct tissue-specific transcriptional response via a PPRE in the li poprotein lipase gene, Embo J 15 (1996) 5336-5348.

PAGE 64

46 [151] J. B. Majithiya, A. N. Paramar, and R. Balaraman, Pioglitazone, a PPARgamma agonist, restores endothelial f unction in aorta of streptozot ocin-induced diabetic rats, Cardiovasc Res 66 (2005) 150-161. [152] A. R. Saltiel, and J. M. Olefsky, Thiazolidinediones in the treatment of insulin resistance and type II diabetes, Diabetes 45 (1996) 1661-1669. [153] N. Kobayashi, T. Ohno, K. Yoshida, H. Fukushima, Y. Mamada, M. Nomura, H. Hirata, Y. Machida, M. Shinoda, N. Suzuki, and H. Matsuoka, Cardioprotective mechanism of telmisartan via PPAR-gamma-e NOS pathway in dahl salt-sensitive hypertensive rats, Am J Hypertens 21 (2008) 576-581. [154] A. R. Collins, W. P. Meehan, U. Kint scher, S. Jackson, S. Wakino, G. Noh, W. Palinski, W. A. Hsueh, and R. E. Law, Troglitazone inhibits formation of early atherosclerotic lesions in diabetic and nondiabetic low density lipoprotein receptordeficient mice, Arterioscler Thromb Vasc Biol 21 (2001) 365-371. [155] J. M. Olefsky, Treatment of insulin resistance with peroxisome proliferatoractivated receptor gamma a gonists, J Clin Invest 106 (2000) 467-472. [156] E. Shinohara, S. Kihara, N. Ouchi, T. Funahashi, T. Nakamura, S. Yamashita, K. Kameda-Takemura, and Y. Matsuzawa, Troglitazone suppresses intimal formation following balloon injury in in sulin-resistant Zucker fatty rats, Atherosclerosis 136 (1998) 275-279. [157] D. S. Calnek, L. Mazzella, S. Roser, J. Roman, and C. M. Hart, Peroxisome proliferator-activated receptor gamma ligands increase release of nitric oxide from endothelial cells, Arterioscl er Thromb Vasc Biol 23 (2003) 52-57. [158] D. H. Cho, Y. J. Choi, S. A. Jo, and I. Jo, Nitric oxide pr oduction and regulation of endothelial nitric-oxide synthase phosphorylation by prolonged treatment with troglitazone: evidence for involvement of pe roxisome proliferator-activated receptor (PPAR) gamma-dependent and PPARgamma-i ndependent signaling pathways, J Biol Chem 279 (2004) 2499-2506. [159] R. Hernandez, T. Teruel, C. de Alvar o, and M. Lorenzo, Rosiglitazone ameliorates insulin resistance in brown adipocytes of Wi star rats by impairi ng TNF-alpha induction of p38 and p42/p44 mitogen-activated protein kinases, Diabetologia 47 (2004) 16151624. [160] S. S. Solomon, L. S. Usdan, and M. R. Palazzolo, Mechanisms involved in tumor necrosis factor-alpha induction of insulin resistance and its reversal by thiazolidinedione(s), Am J Med Sci 322 (2001) 75-78.

PAGE 65

47 [161] K. Goya, S. Sumitani, M. Otsuki, X. Xu, H. Yamamoto, S. Kurebayashi, H. Saito, H. Kouhara, and S. Kasayama, The thiazolidinedione drug troglitazone up-regulates nitric oxide synthase expression in vascular e ndothelial cells, J Diab etes Complications 20 (2006) 336-342. [162] B. L. Goodwin, K. D. Corbin, L. C. Pendleton, M. M. Levy, L. P. Solomonson, and D. C. Eichler, Troglitazone up-regulates vasc ular endothelial argininosuccinate synthase, Biochem Biophys Res Commun 370 (2008) 254-258. [163] U. Scherrer, D. Randin, P. Vollenweide r, L. Vollenweider, and P. Nicod, Nitric oxide release accounts for insulin's vascular effects in humans, J Clin Invest 94 (1994) 2511-2515. [164] H. O. Steinberg, G. Br echtel, A. Johnson, N. Fineberg, and A. D. Baron, Insulinmediated skeletal muscle vasodilation is ni tric oxide dependent. A novel action of insulin to increase nitric oxide release, J Clin Invest 94 (1994) 1172-1179. [165] K. Kuboki, Z. Y. Jiang, N. Takahara, S. W. Ha, M. Igarashi, T. Yamauchi, E. P. Feener, T. P. Herbert, C. J. Rhodes, and G. L. King, Regulation of endothelial constitutive nitric oxide synt hase gene expression in e ndothelial cells and in vivo : a specific vascular action of insulin, Circulation 101 (2000) 676-681. [166] B. Fisslthaler, T. Benzing, R. Busse, and I. Fleming, Insulin enhances the expression of the endothelial nitric oxide synthase in native endothelial cells: a dual role for Akt and AP-1, Nitric Oxide 8 (2003) 253-261. [167] G. A. Spinas, R. Laffranchi, I. Franc oys, I. David, C. Richter, and M. Reinecke, The early phase of glucose-stimulated in sulin secretion requires nitric oxide, Diabetologia 41 (1998) 292-299. [168] M. Nakata, and T. Yada, Endocrinology: nitric oxide-mediated insulin secretion in response to citrulline in is let beta-cells, Pancreas 27 (2003) 209-213. [169] T. Kobayashi, and K. Kamata, Short-te rm insulin treatment and aortic expressions of IGF-1 receptor and VEGF mR NA in diabetic rats, Am J Physiol Heart Circ Physiol 283 (2002) H1761-1768. [170] S. Oyadomari, T. Gotoh, K. Aoyagi E. Araki, M. Shichiri, and M. Mori, Coinduction of endothelial nitric oxide syntha se and arginine recycling enzymes in aorta of diabetic rats, Nitric Oxide 5 (2001) 252-260. [171] S. Kliche, and J. Waltenberger, VEGF receptor signaling and endothelial function, IUBMB Life 52 (2001) 61-66.

PAGE 66

48 [172] J. D. Hood, C. J. Meininger, M. Zich e, and H. J. Granger, VEGF upregulates ecNOS message, protein, and NO production in human endothelial ce lls, Am J Physiol 274 (1998) H1054-1058. [173] A. Bouloumie, V. B. Sc hini-Kerth, and R. Busse, Va scular endothelial growth factor up-regulates nitric oxide synthase expression in endo thelial cells, Cardiovasc Res 41 (1999) 773-780. [174] T. Teruel, R. Hernandez, and M. Lorenz o, Ceramide mediates insulin resistance by tumor necrosis factor-alpha in brown adi pocytes by maintaining Akt in an inactive dephosphorylated state, Diabetes 50 (2001) 2563-2571. [175] P. Der, J. Cui, and D. K. Das, Role of lipid rafts in ceramide and nitric oxide signaling in the ischemic and preconditioned hearts, J Mol Cell Cardiol 40 (2006) 313320. [176] G. Garcia-Cardena, R. Fa n, D. F. Stern, J. Liu, and W. C. Sessa, Endothelial nitric oxide synthase is regulated by tyrosine phos phorylation and interacts with caveolin-1, J Biol Chem 271 (1996) 27237-27240. [177] G. R. Hellermann, B. R. Flam, D. C. Ei chler, and L. P. Solomonson, Stimulation of receptor-mediated nitric oxide production by va nadate, Arterioscler Thromb Vasc Biol 20 (2000) 2045-2050. [178] K. K. Wu, Regulation of endothelial nitric oxide synthase activity and gene expression, Ann N Y Acad Sci 962 (2002) 122-130. [179] R. C. Venema, Post-trans lational mechanisms of endot helial nitric oxide synthase regulation by bradykinin, Int Immunopharmacol 2 (2002) 1755-1762. [180] D. M. Dudzinski, and T. Michel, Life history of eNOS: partners and pathways, Cardiovasc Res 75 (2007) 247-260. [181] C. Li, L. Ruan, S. G. Sood, A. Papape tropoulos, D. Fulton, and R. C. Venema, Role of eNOS phosphorylation at Ser-116 in regulation of eNOS ac tivity in endothelial cells, Vascul Pharmacol 47 (2007) 257-264. [182] B. G. Drew, N. H. Fi dge, G. Gallon-Beaumier, B. E. Kemp, and B. A. Kingwell, High-density lipoprotein and apo lipoprotein AI increase endothelial NO synthase activity by protein association and multisite phosphor ylation, Proc Natl Acad Sci U S A 101 (2004) 6999-7004.

PAGE 67

49 [183] B. J. Michell, Z. Chen, T. Tiganis, D. Stapleton, F. Katsis, D. A. Power, A. T. Sim, and B. E. Kemp, Coordinated control of endothelial nitric-oxide synthase phosphorylation by protein kinase C and the cAMP-dependent protein kinase, J Biol Chem 276 (2001) 17625-17628. [184] I. Fleming, B. Fisslthaler, S. Dimmeler, B. E. Kemp, and R. Busse, Phosphorylation of Thr(495) regulates Ca(2+) /calmodulin-dependent endothelial nitric oxide synthase activity, Circ Res 88 (2001) E68-75. [185] M. B. Harris, H. Ju, V. J. Venema, H. Liang, R. Zou, B. J. Michell, Z. P. Chen, B. E. Kemp, and R. C. Venema, Reciprocal phos phorylation and regula tion of endothelial nitric-oxide synthase in response to bradykinin stimulation, J Biol Chem 276 (2001) 16587-16591. [186] C. Partovian, Z. Zhuang, K. Moodie, M. Lin, N. Ouchi, W. C. Sessa, K. Walsh, and M. Simons, PKCalpha activates eNOS and incr eases arterial blood flow in vivo, Circ Res 97 (2005) 482-487. [187] C. Rask-Madsen, and G. L. King, Di fferential regulation of VEGF signaling by PKC-alpha and PKC-epsilon in endothelial ce lls, Arterioscler Thromb Vasc Biol 28 (2008) 919-924. [188] J. B. Michel, O. Feron, K. Sase, P. Prabhakar, and T. Michel, Caveolin versus calmodulin. Counterbalancing allosteric modulat ors of endothelial nitric oxide synthase, J Biol Chem 272 (1997) 25907-25912. [189] B. J. Michell, M. B. Harris, Z. P. Chen, H. Ju, V. J. Venema M. A. Blackstone, W. Huang, R. C. Venema, and B. E. Kemp, Identification of re gulatory sites of phosphorylation of the bovine e ndothelial nitric-oxide synthase at serine 617 and serine 635, J Biol Chem 277 (2002) 42344-42351. [190] Y. C. Boo, J. Hwang, M. Sykes, B. J. Michell, B. E. Kemp, H. Lum, and H. Jo, Shear stress stimulates phosphorylation of eNOS at Ser(635) by a protein kinase Adependent mechanism, Am J P hysiol Heart Circ Physiol 283 (2002) H1819-1828. [191] A. Brouet, P. Sonveaux, C. Dessy, J. L. Balligand, and O. Feron, Hsp90 ensures the transition from the early Ca2+-dependent to the late phosphorylation-dependent activation of the endothelial nitric-oxide syntha se in vascular endothe lial growth factorexposed endothelial cells, J Biol Chem 276 (2001) 32663-32669. [192] S. W. Bae, H. S. Kim, Y. N. Cha, Y. S. Park, S. A. Jo, and I. Jo, Rapid increase in endothelial nitric oxide pr oduction by bradykinin is medi ated by protein kinase A signaling pathway, Bioche m Biophys Res Commun 306 (2003) 981-987.

PAGE 68

50 [193] P. M. Bauer, D. Fulton, Y. C. Boo, G. P. Sorescu, B. E. Kemp, H. Jo, and W. C. Sessa, Compensatory phosphorylation and protei n-protein interactions revealed by loss of function and gain of function mutants of multiple serine phosphorylation sites in endothelial nitric-oxide s ynthase, J Biol Chem 278 (2003) 14841-14849. [194] Y. C. Boo, H. J. Kim, H. Song, D. Fulton, W. Sessa, and H. Jo, Coordinated regulation of endothelial nitric oxide synthase activity by phosphorylation and subcellular localization, Free Radic Biol Med 41 (2006) 144-153. [195] H. Cai, Z. Li, M. E. Davis, W. Kanner, D. G. Harrison, and S. C. Dudley, Jr., Aktdependent phosphorylation of serine 1179 and mitogen-ac tivated protein kinase kinase/extracellular sign al-regulated kinase 1/2 cooperatively medi ate activation of the endothelial nitric-oxide synthase by hydrogen peroxide, Mol Pharmacol 63 (2003) 325331. [196] Z. P. Chen, K. I. Mitchelhill, B. J. Michell, D. Stapleton, I. Rodriguez-Crespo, L. A. Witters, D. A. Power, P. R. Ortiz de Montellano, and B. E. Kemp, AMP-activated protein kinase phosphorylation of endot helial NO synthase, FEBS Lett 443 (1999) 285289. [197] D. Feliers, X. Chen, N. Akis, G. G. Choudhury, M. Madaio, and B. S. Kasinath, VEGF regulation of endothelial nitric oxide synthase in glomerular endothelial cells, Kidney Int 68 (2005) 1648-1659. [198] D. Fulton, J. P. Gratton, T. J. McCabe, J. Fontana, Y. Fujio, K. Walsh, T. F. Franke, A. Papapetropoulos, and W. C. Sessa Regulation of endothelium-derived nitric oxide production by the protei n kinase Akt, Nature 399 (1999) 597-601. [199] B. Gallis, G. L. Corthals, D. R. Goodle tt, H. Ueba, F. Kim, S. R. Presnell, D. Figeys, D. G. Harrison, B. C. Berk, R. Aebersold, and M. A. Corson, Identification of flow-dependent endothelial nitric-oxi de synthase phosphorylation sites by mass spectrometry and regulation of phosphoryla tion and nitric oxide production by the phosphatidylinositol 3-kinase i nhibitor LY294002, J Biol Chem 274 (1999) 3010130108. [200] D. S. Gelinas, P. N. Bernatchez, S. Rollin, N. G. Bazan, and M. G. Sirois, Immediate and delayed VEGF-mediated NO synthe sis in endothelial cells: role of PI3K, PKC and PLC pathways, Br J Pharmacol 137 (2002) 1021-1030. [201] M. Montagnani, H. Chen, V. A. Barr, and M. J. Quon, Insulin-stimulated activation of eNOS is independent of Ca2+ but requi res phosphorylation by Ak t at Ser(1179), J Biol Chem 276 (2001) 30392-30398.

PAGE 69

51 [202] J. A. Reihill, M. A. Ewart, D. G. Hardie, and I. P. Salt, AMP-activated protein kinase mediates VEGF-stimulated endot helial NO production, Biochem Biophys Res Commun 354 (2007) 1084-1088. [203] Y. C. Boo, Shear stress stimulates phos phorylation of protein kinase A substrate proteins including endothelial ni tric oxide synthase in endo thelial cells, Exp Mol Med 38 (2006) 63-71. [204] K. Imami, N. Sugiyama, Y. Kyono, M. Tomita, and Y. Ishihama, Automated phosphoproteome analysis for cultured can cer cells by two-dimensional nanoLC-MS using a calcined titania/C18 biphasic column, Anal Sci 24 (2008) 161-166. [205] P. Lane, G. Hao, and S. S. Gross, Snitrosylation is emerging as a specific and fundamental posttranslational protein modificatio n: head-to-head comparison with Ophosphorylation, Sci STKE 2001 (2001) RE1. [206] P. A. Erwin, A. J. Lin, D. E. Golan, and T. Michel, Recept or-regulated dynamic Snitrosylation of endothelial nitric-oxide synt hase in vascular endot helial cells, J Biol Chem 280 (2005) 19888-19894. [207] L. J. Robinson, and T. Michel, Mutagenesis of palmitoylation sites in endothelial nitric oxide synthase identifies a novel motif for dual acylation and subcellular targeting, Proc Natl Acad Sci U S A 92 (1995) 11776-11780. [208] E. Gonzalez, R. Kou, A. J. Lin, D. E. Golan, and T. Michel, Subcellular targeting and agonist-induced site-specifi c phosphorylation of endothelia l nitric-oxide synthase, J Biol Chem 277 (2002) 39554-39560. [209] P. W. Shaul, E. J. Smart, L. J. Robi nson, Z. German, I. S. Yuhanna, Y. Ying, R. G. Anderson, and T. Michel, Acylation targets endothelial n itric-oxide synthase to plasmalemmal caveolae, J Biol Chem 271 (1996) 6518-6522. [210] J. Liu, G. Garcia-Cardena, and W. C. Sessa, Palmitoylation of endothelial nitric oxide synthase is necessary for optimal stimulat ed release of nitric oxide: implications for caveolae localization, Biochemistry 35 (1996) 13277-13281. [211] N. Fulop, R. B. Marchase, and J. C. Chatham, Role of pr otein O-linked N-acetylglucosamine in mediating cell function and survival in the cardiovascular system, Cardiovasc Res 73 (2007) 288-297. [212] F. I. Comer, and G. W. Hart, O-Gl cNAc and the control of gene expression, Biochim Biophys Acta 1473 (1999) 161-171.

PAGE 70

52 [213] M. Federici, R. Menghini A. Mauriello, M. L. Hribal F. Ferrelli, D. Lauro, P. Sbraccia, L. G. Spagnoli, G. Sesti, and R. Lauro, Insulin-dependent activation of endothelial nitric oxide synt hase is impaired by O-linked gl ycosylation modification of signaling proteins in human coronary endothelial cells, Circulation 106 (2002) 466-472. [214] B. Musicki, M. F. Kramer, R. E. B ecker, and A. L. Burnett, Inactivation of phosphorylated endothelial nitric oxide syntha se (Ser-1177) by O-Gl cNAc in diabetesassociated erectile dysfunction, Proc Natl Acad Sci U S A 102 (2005) 11870-11875. [215] C. Brasse-Lagnel, A. Fairand, A. La voinne, and A. Husson, Glutamine stimulates argininosuccinate synthetase gene expression through cytosolic O-glycosylation of Sp1 in Caco-2 cells, J Biol Chem 278 (2003) 52504-52510. [216] T. M. Vondriska, J. M. Pass, and P. Ping, Scaffold proteins and assembly of multiprotein signaling complexes, J Mol Cell Cardiol 37 (2004) 391-397. [217] P. W. Shaul, Regulation of endothelial nitric oxide syntha se: location, location, location, Annu Rev Physiol 64 (2002) 749-774. [218] M. S. Goligorsky, H. Li, S. Brodsky, and J. Chen, Relationships between caveolae and eNOS: everything in proximity and the pr oximity of everything, Am J Physiol Renal Physiol 283 (2002) F1-10. [219] Q. Zhang, J. E. Church, D. Jagnandan, J. D. Catravas, W. C. Sessa, and D. Fulton, Functional relevance of Golgiand plasma membrane-localized endothelial NO synthase in reconstituted endothelial cells, Arterioscler Thromb Vasc Biol 26 (2006) 1015-1021. [220] D. Fulton, J. Fontana, G. Sowa, J. P. Gratton, M. Lin, K. X. Li, B. Michell, B. E. Kemp, D. Rodman, and W. C. Sessa, Localizat ion of endothelial nitric-oxide synthase phosphorylated on serine 1179 and nitric oxide in Golgi and plasma membrane defines the existence of two pools of active enzyme, J Biol Chem 277 (2002) 4277-4284. [221] X. A. Li, W. Everson, and E. J. Sm art, Nitric oxide, caveolae, and vascular pathology, Cardiovasc Toxicol 6 (2006) 1-13. [222] F. A. Sanchez, N. B. Savalia, R. G. Du ran, B. K. Lal, M. P. Boric, and W. N. Duran, Functional significance of differential eNOS transl ocation, Am J Physiol Heart Circ Physiol 291 (2006) H1058-1064. [223] S. Mukhopadhyay, F. Xu, and P. B. Sehgal, Aberrant cytoplasmi c sequestration of eNOS in endothelial cells after monocrota line, hypoxia, and senescence: live-cell caveolar and cytoplasmic NO imaging, Am J Physiol Heart Circ Physiol 292 (2007) H1373-1389.

PAGE 71

53 [224] D. Jagnandan, W. C. Sessa, and D. Fult on, Intracellular location regulates calciumcalmodulin-dependent activation of organelle-re stricted eNOS, Am J Physiol Cell Physiol 289 (2005) C1024-1033. [225] J. E. Church, and D. Fulton, Differences in eNOS activity because of subcellular localization are dictated by phosphorylation state rather than the local calcium environment, J Biol Chem 281 (2006) 1477-1488. [226] L. Loufrani, and D. He nrion, Role of the cytoskeleton in flow (shear stress)induced dilation and remodeling in resistance arteries Med Biol Eng Comput 46 (2008) 451-460. [227] Y. Su, D. Kondrikov, and E. R. Block, Beta-actin: a regulator of NOS-3, Sci STKE 2007 (2007) pe52. [228] Y. Hiroi, Z. Guo, Y. Li, A. H. Be ggs, and J. K. Liao, Dynamic regulation of endothelial NOS mediated by competitive interaction with al pha-actinin-4 and calmodulin, Faseb J 22 (2008) 1450-1457. [229] R. R. Sprenger, R. D. Fontijn, J. va n Marle, H. Pannekoek, and A. J. Horrevoets, Spatial segregation of transport and signa lling functions between human endothelial caveolae and lipid raft proteomes, Biochem J 400 (2006) 401-410. [230] Y. Su, S. I. Zharikov, and E. R. Bloc k, Microtubule-active agents modify nitric oxide production in pulmonary artery endot helial cells, Am J Physiol Lung Cell Mol Physiol 282 (2002) L1183-1189. [231] K. Schilling, N. Opitz, A. Wiesenthal, S. Oess, R. Tikkanen, W. Muller-Esterl, and A. Icking, Translocation of endothelial nitric -oxide synthase involves a ternary complex with caveolin-1 and NOSTRIN, Mol Biol Cell 17 (2006) 3870-3880. [232] K. Zimmermann, N. Opitz, J. Dedio, C. Renne, W. Muller-Esterl, and S. Oess, NOSTRIN: a protein modulating nitric oxide release and subcellu lar distribution of endothelial nitric oxide syntha se, Proc Natl Acad Sci U S A 99 (2002) 17167-17172. [233] J. Dedio, P. Konig, P. Wohlfart, C. Sc hroeder, W. Kummer, and W. Muller-Esterl, NOSIP, a novel modulator of endothelial ni tric oxide synthase activity, Faseb J 15 (2001) 79-89. [234] M. Schleicher, F. Brundin, S. Gross, W. Muller-Esterl, and S. Oess, Cell cycleregulated inactivation of endot helial NO synthase through NO SIP-dependent targeting to the cytoskeleton, Mol Cell Biol 25 (2005) 8251-8258.

PAGE 72

54 [235] H. Ju, R. Zou, V. J. Venema, and R. C. Venema, Direct intera ction of endothelial nitric-oxide synthase and caveolin-1 inhi bits synthase activity, J Biol Chem 272 (1997) 18522-18525. [236] P. Prabhakar, H. S. Thatte, R. M. Goetz, M. R. Cho, D. E. Golan, and T. Michel, Receptor-regulated translocati on of endothelial nitric-oxide synthase, J Biol Chem 273 (1998) 27383-27388. [237] V. Rizzo, D. P. McIntosh, P. Oh, and J. E. Schnitzer, In situ flow activates endothelial nitric oxide syntha se in luminal caveolae of endo thelium with rapid caveolin dissociation and calmodulin association, J Biol Chem 273 (1998) 34724-34729. [238] H. Ju, V. J. Venema, M. B. Marrero, a nd R. C. Venema, Inhibitory interactions of the bradykinin B2 receptor with endothelial nitric-oxide synthase, J Biol Chem 273 (1998) 24025-24029. [239] G. Garcia-Cardena, R. Fan, V. Shah, R. Sorrentino, G. Cirino, A. Papapetropoulos, and W. C. Sessa, Dynamic activation of endothelial nitric oxide synthase by Hsp90, Nature 392 (1998) 821-824. [240] S. Takahashi, and M. E. Mendelsohn, Synergistic activation of endothelial nitricoxide synthase (eNOS) by HSP90 and Ak t: calcium-independent eNOS activation involves formation of an HSP90-Akt-CaM-bound eNOS complex, J Biol Chem 278 (2003) 30821-30827. [241] J. P. Gratton, J. Fontana, D. S. O' Connor, G. Garcia-Cardena, T. J. McCabe, and W. C. Sessa, Reconstitution of an endothelial nitric-oxide synthase (eNOS), hsp90, and caveolin-1 complex in vitro. Evidence th at hsp90 facilitates calmodulin stimulated displacement of eNOS from caveolin-1, J Biol Chem 275 (2000) 22268-22272. [242] J. B. Michel, O. Feron, D. Sacks, and T. Michel, Reciprocal regulation of endothelial nitric-oxide s ynthase by Ca2+-calmodulin and caveolin, J Biol Chem 272 (1997) 15583-15586. [243] C. Li, W. Huang, M. B. Harris, J. M. Goolsby, and R. C. Venema, Interaction of the endothelial nitric oxide synthase with the CAT-1 arginine transporter enhances NO release by a mechanism not involving arginine transport, Biochem J 386 (2005) 567-574. [244] Y. Zhao, J. Zhang, H. Li, Y. Li, J. Ren, M. Luo, and X. Zheng, An NADPH sensor protein (HSCARG) down-regul ates nitric oxide synthe sis by association with argininosuccinate synthetase a nd is essential for epithelial ce ll viability, J Biol Chem 283 (2008) 11004-11013.

PAGE 73

55 SPECIFIC AIMS Purpose Heart disease is the number one killer of Americans. As the incidence of obesity, diabetes and metabolic syndrome continues rising at an alarming rate, so will the prevalence of endothelial dysfunction [1] Since endothelial dysfunction is often accompanied by diminished or excessive nitric oxide (NO) production [2], it is essential to continue our efforts to gain a better unde rstanding of the regulation of NO synthesis. Our studies were designed to expand our global understanding of vascular biology by assessing the regulation of th e citrulline-NO cycle from a different perspective. Since much of the focus has been on the regulation of endothelial nitric oxide synthase (eNOS), our research is targeted at a better understanding of another equally important component of the citrulline-NO cycle, argi ninosuccinate synthase (AS). Central Question and Hypothesis AS is an enzyme that is important for the production of nitric oxide. The central question we are addressing with the work de scribed is: How does AS regulate endothelial NO production? Due to the fundamental role of AS for vascular biology, we hypothesize

PAGE 74

56 that multiple mechanisms regulate AS function to control NO production and support endothelial function. Specific Aim 1 The role of eNOS in the control of NO production is well studied [2]. Much less is known about the regulatory role of AS. We and others have demonstrated the importance of AS expression for the production of NO [3-14]. We hypothesize that a multitude of stimuli alter AS expressi on to enhance NO production in a manner consistent with the substrate needs of eNOS. In Specific Aim 1, described in Chapter One, we will first examine the role of AS overexpression in regulating NO production. We will then determine whether insulin, vasc ular endothelial growth factor (VEGF) and ceramide impact AS expression coordinately with eNOS. Specific Aim 2 Although regulation at the leve l of transcription and tran slation is an important mechanism controlling the level of function of an enzyme, this type of regulation seldom accounts for acute changes in enzymatic activit y. NO synthesis is a constant and dynamic process. It is well documented that eNOS is regulated by a co mplex set of reversible posttranslational modifications [1517]. Since AS is the source of substrate for eNOS [9, 11], we hypothesize that a similar pattern of posttranslational modifications exists for the acute regulation of AS function. In Specific Aim 2, described in Chapter Two, we will

PAGE 75

57 determine whether AS is an endogenous phosphoprotein, define the biological significance of AS phosphoryl ation, and uncover the possible mechanisms by which AS phosphorylation regulates its function. Specific Aim 3 Post-translational regulation of protei n function often involves subcellular localization and dynamic protei n interactions. There is a complex literature surrounding the regulation of eNOS trafficking, activa tion and function that is driven by protein interactions [17-19]. However, very little is known about such regulatory mechanisms controlling AS function. In Specific Aim 3, described in Chapter Three, we will define the subcellular localization of AS and identify key interacting partners. Working Model It is our belief that th e regulation of caveolae-locali zed AS, in conjunction with eNOS, is mediated by an interre lated set of mechanisms that controls the expression, post translational modifications and protein interactions that are so critical for the overall function of the system. As demonstrated in Figure 1, we hypothesize that agonists and antagonists will coordinately re gulate the expression, activation and protein interactions of AS and eNOS so that the level of NO produ ced is adequate to meet the current needs of the cell. In addition, it is our belief that the multiple mechanisms that regulate AS and eNOS are inter-related, yet sp ecific to environmental si gnals. Although there might be

PAGE 76

58 instances where there is some discoordinate regulation of AS and eNOS, such instances are necessary in scenarios where an initial change in regulation of one of these two enzymes would signal a need for a change in NO production which would then be followed by a coordinate regulation of the other enzyme for the express purpose of restoring homeostasis. We also believe that there are specific mechanisms that control the regulation of the caveolae-localized citrulline-NO complex in comparison to other locations within the cell. Importantly, the coordinate, multi-level regulation of AS and eNOS is designed specifically to regulate th e synthesis of NO in an attempt to support the health of the endothelium and ultimately of the cardiovascular system. Any derangements of the physiological regulation of this complex, even if minor, would have a great impact on vascular health and explains the prominence of vascular di sorders seen with a variety of metabolic imbalances.

PAGE 77

Figure 1: Working Model of the Regulati on of the Citrulline-NO Cycle Under Physiological Conditions. Figure depicts the central components of our study of the citrulline-NO cycle, AS and eNOS, as a caveol ae-localized complex that is coordinately regulated by agonists and antagonists via expression, phosphorylation and protein interactions in an effort to produce adequate amounts of NO to promote vascular health. 59

PAGE 78

60 References [1] W. Rosamond, K. Flegal, G. Friday, K. Fu rie, A. Go, K. Greenlund, N. Haase, M. Ho, V. Howard, B. Kissela, S. Kittner, D. Lloyd-Jones, M. McDermott, J. Meigs, C. Moy, G. Nichol, C. J. O'Donnell, V. Roger, J. Rumsfeld, P. Sorlie, J. Steinberger, T. Thom, S. Wasserthiel-Smoller, and Y. Hong, Heart dise ase and stroke sta tistics--2007 update: a report from the American Heart Association Statistics Committee and Stroke Statistics Subcommittee, Circulation 115 (2007) e69-171. [2] U. Forstermann, and T. Munzel, Endothelial nitric oxide synthase in vascular disease: from marvel to menace, Circulation 113 (2006) 1708-1714. [3] B. L. Goodwin, K. D. Corbin, L. C. Pendleton, M. M. Levy, L. P. Solomonson, and D. C. Eichler, Troglitazone up-regulates vasc ular endothelial argininosuccinate synthase, Biochem Biophys Res Commun 370 (2008) 254-258. [4] B. L. Goodwin, L. C. Pe ndleton, M. M. Levy, L. P. Solomonson, and D. C. Eichler, Tumor necrosis factor-alpha reduces argininosuccinate synthase expression and nitric oxide production in aortic endothelial cel ls, Am J Physiol Heart Circ Physiol 293 (2007) H1115-1121. [5] B. L. Goodwin, L. P. So lomonson, and D. C. Eichler, Argininosuccinate synthase expression is required to maintain nitric ox ide production and cell viability in aortic endothelial cells, J Biol Chem 279 (2004) 18353-18360. [6] B. R. Flam, D. C. Eichler, and L. P. Solomonson, Endothelial nitric oxide production is tightly coupled to the citrul line-NO cycle, Nitric Oxide 17 (2007) 115-121. [7] B. R. Flam, P. J. Hartmann, M. Harrell-B ooth, L. P. Solomonson, and D. C. Eichler, Caveolar localization of argi nine regeneration enzymes, ar gininosuccinate synthase, and lyase, with endothelial nitric oxide synthase, Nitric Oxide 5 (2001) 187-197. [8] L. Xie, and S. S. Gross, Argininosucci nate synthetase overe xpression in vascular smooth muscle cells poten tiates immunostimulant-induced NO production, J Biol Chem 272 (1997) 16624-16630. [9] L. Xie, Y. Hattori, N. Tume, and S. S. Gross, The preferred source of arginine for high-output nitric oxide synthesis in blood vessels, Semin Perinatol 24 (2000) 42-45. [10] C. W. Shuttleworth, A. J. Burns, S. M. Ward, W. E. O'Brien, and K. M. Sanders, Recycling of L-citrulline to sustain nitric oxide-dependent enteric neurotransmission, Neuroscience 68 (1995) 1295-1304.

PAGE 79

61 [11] L. J. Shen, K. Beloussow, and W. C. Shen, Accessibility of endothelial and inducible nitric oxide synthase to the intracellula r citrulline-arginine regeneration pathway, Biochem Pharmacol 69 (2005) 97-104. [12] S. Oyadomari, T. Gotoh, K. Aoyagi E. Araki, M. Shichiri, and M. Mori, Coinduction of endothelial nitric oxide syntha se and arginine recycling enzymes in aorta of diabetic rats, Nitric Oxide 5 (2001) 252-260. [13] T. Koga, W. Y. Zhang, T. Gotoh, S. Oyadomari, H. Tanihara, and M. Mori, Induction of citrulline-nit ric oxide (NO) cycle en zymes and NO production in immunostimulated rat RPE-J cells, Exp Eye Res 76 (2003) 15-21. [14] G. Hao, L. Xie, and S. S. Gross, Argininosuccinate synthe tase is reversibly inactivated by S-nitrosylation in vi tro and in vivo, J Biol Chem 279 (2004) 36192-36200. [15] R. Govers, and T. J. Rabelink, Cellular regulation of endothelial nitric oxide synthase, Am J Physiol Renal Physiol 280 (2001) F193-206. [16] R. C. Venema, Post-translational mechan isms of endothelial nitric oxide synthase regulation by bradykinin, Int Immunopharmacol 2 (2002) 1755-1762. [17] D. M. Dudzinski, and T. Michel, Life history of eNOS: partners and pathways, Cardiovasc Res 75 (2007) 247-260. [18] P. W. Shaul, Regulati on of endothelial nitric oxide synthase: location, location, location, Annu Rev Physiol 64 (2002) 749-774. [19] M. S. Goligorsky, H. Li, S. Brodsky, a nd J. Chen, Relationships between caveolae and eNOS: everything in proximity and the pr oximity of everything, Am J Physiol Renal Physiol 283 (2002) F1-10.

PAGE 80

62 CHAPTER ONE ARGININOSUCCINATE SYNTHASE FUNCTION AND EXPRESSION Overview Argininosuccinate synthase (A S) is a key, regulated step of the citrulline-nitric oxide (NO) cycle. We have pr eviously demonstrated several levels of regulation of AS expression that impact NO production. First, when AS expression is reduced utilizing siRNA, NO production, AS enzymatic activ ity and endothelial cell viability is diminished. The loss of endothelial cells is via apoptosis and can be rescued with an NO donor, demonstrating the direct role of AS activity and NO production in maintaining endothelial cell viabilit y. In addition, TNF a cytokine associated with vascular disease, leads to decreased expression of AS unde r conditions of chronic inflammation. This decreased expression occurs via a reduction in the ability of SP-1 elements to activate the AS proximal promoter, similar to the effects of chronic TNF treatment on the eNOS promoter. Finally, we have demonstrated that AS expression is increased by the PPAR agonist drug, troglitazone, due to enhanced activity at a distal AS PPAR responsive element (PPRE). To expand our understandi ng of AS transcriptional regulation, we explored several pathways. First, we test ed whether AS overexpression in endothelial cells enhances NO production. Our data demonstrates that AS overexpression

PAGE 81

63 significantly increased basal NO production within 6 hours of transfection. Since insulin is a key regulator of vasod ilation and vascular function, the hypothesis that insulin enhances AS expression and NO production wa s tested. Western blot and real time PCR experiments showed that AS protein a nd mRNA expression were up-regulated by physiological doses of insulin. Insulin c oordinately regulated eNOS expression. Luciferase assay data also s uggested that insulin may be ac tivating a distal AS promoter element. Nitric oxide assays demonstrated that insulin alone can increase NO production and that it also acts synergistically with bradykinin and the calcium ionophore A23187 to increase NO production. This suggests that insu lin up-regulation of AS is part of the mechanism by which it increases NO produc tion. Since VEGF mediates important endothelium-specific functions such as angi ogenesis and vasodilation, the hypothesis that VEGF enhances AS expression and NO producti on was examined. Western blot analysis demonstrated that VEGF increases AS expression. We also demonstrated a timedependent increase in NO production in respons e to VEGF. Finally, we investigated the effects of ceramide on AS and eNOS expres sion. Ceramide is a bioactive sphingolipid with roles in cell si gnaling and apoptosis. Often times, the pathogenic effects of TNF are mediated by ceramide. This prompted th e hypothesis that cera mide would diminish AS expression under pathogenic conditions. Western blot analyses demonstrated that ceramide decreased AS and eNOS expression. In addition, ceramide diminished eNOS activation. Collectively, the data in this Chapter suggests th at AS expression is highly regulated in a manner that is c onsistent with its role in suppo rting the catalytic activity of eNOS.

PAGE 82

64 Materials and Methods Bovine Endothelial Cell Culture: Bovine aortic endotheli al cells (BAEC) were isolated by our laboratory from bovine aort a following the procedure of Gospodarowicz et al [1], and were used from passages 4-10. The study of bovine AS is supported by the extensive use of BAEC in rese arch and the fact that AS shares a high sequence identity between human [2], bovine [3], rat [4] and mouse [5]. BAEC were cultured at 37 C under an atmosphere of 5% CO2 in complete Dulbeccos Modified Eagles Medium (DMEM) (1 g/L glucose, Mediatech) that contained 10% fetal bovi ne serum (Hyclone Laboratories), 100 units/ml penicillin and 100 g/ml streptomycin (Mediatech). Cells were treated once they reached confluence. AS Expression Vector: The AS plasmid contains both a V5 and 6X-His tag at its C-terminus. The vector map and sequence are below:

PAGE 83

Figure 2: AS Expression Vector Map -43bAS_pcDNA3.1V5HisB (updated 04Aug05)6770 bps 1000 2000 3000 4000 5000 6000 BglII NdeI KpnI BamHI FseI BsaBI XhoI BsrGI EcoRI EcoRV BstXI NotI XhoI XbaI SacII PmeI BsaBI BssHII PvuI ScaI AS Insert AS start V5 6xHis 65

PAGE 84

Figure 3: AS Expression Vector Sequence. Green = AS start; Red = AS stop; Turquoise = codons mutated to make T131, S180 and S189 variants in Chapter Two. 1 GACGGATCGG GAGATCTCCC GATCCCCTAT GGTGCACTCT CAGTACAATC TGCTCTGATG 61 CCGCATAGTT AAGCCAGTAT CTGCTCCCTG CTTGTGTGTT GGAGGTCGCT GAGTAGTGCG 121 CGAGCAAAAT TTAAGCTACA ACAAGGCAAG GCTTGACCGA CAATTGCATG AAGAATCTGC 181 TTAGGGTTAG GCGTTTTGCG CTGCTTCGCG ATGTACGGGC CAGATATACG CGTTGACATT 241 GATTATTGAC TAGTTATTAA TAGTAATCAA TTACGGGGTC ATTAGTTCAT AGCCCATATA 301 TGGAGTTCCG CGTTACATAA CTTACGGTAA ATGGCCCGCC TGGCTGACCG CCCAACGACC 361 CCCGCCCATT GACGTCAATA ATGACGTATG TTCCCATAGT AACGCCAATA GGGACTTTCC 421 ATTGACGTCA ATGGGTGGAG TATTTACGGT AAACTGCCCA CTTGGCAGTA CATCAAGTGT 481 ATCATATGCC AAGTACGCCC CCTATTGACG TCAATGACGG TAAATGGCCC GCCTGGCATT 541 ATGCCCAGTA CATGACCTTA TGGGACTTTC CTACTTGGCA GTACATCTAC GTATTAGTCA 601 TCGCTATTAC CATGGTGATG CGGTTTTGGC AGTACATCAA TGGGCGTGGA TAGCGGTTTG 661 ACTCACGGGG ATTTCCAAGT CTCCACCCCA TTGACGTCAA TGGGAGTTTG TTTTGGCACC 721 AAAATCAACG GGACTTTCCA AAATGTCGTA ACAACTCCGC CCCATTGACG CAAATGGGCG 781 GTAGGCGTGT ACGGTGGGAG GTCTATATAA GCAGAGCTCT CTGGCTAACT AGAGAACCCA 841 CTGCTTACTG GCTTATCGAA ATTAATACGA CTCACTATAG GGAGACCCAA GCTGGCTAGT 901 TAAGCTTGGT ACCGAGCTCG GATCCGCCCT GCTCCGCCGA CTGCTGCCGC CGCTGGTCAC 961 CCGTCACGAT GTCCGGCAAA GGCTCCGTGG TTCTGGCCTA CAGTGGGGGC CTGGACACCT 1021 CCTGCATCCT CGTGTGGCTG AAGGAGCAAG GCTATGACGT CATTGCCTAC CTGGCCAACA 1081 TCGGCCAGAA AGAAGACTTT GAGGAAGCCA GGAAGAAGGC GCTGAAGCTT GGGGCCAAAA 1141 AGGTGTTCAT TGAGGACATC AGCAAGGAGT TTGTGGAGGA GTTCATCTGG CCGGCCATCC 1201 AGTCCAGCGC ACTGTACGAG GACCGATACC TCCTGGGCAC CTCTCTCGCC AGGCCCTGCA 1261 TCGCCCGCAA GCAGGTGGAG ATCGCCCAGC GAGAAGGAGC CAAGTATGTG TCTCACGGCG 1321 CCACAGGAAA GGGGAACGAC CAGATCCGGT TTGAGCTCAC CTGCTACTCG CTGGCCCCAC 1381 AGATCAAGGT CATCGCTCCC TGGAGGATGC CCGAGTTCTA TAACCGCTTC CAGGGCCGCA 1441 ACGATCTGAT GGAGTATGCG AAGCAACATG GAATCCCCGT CCCAGTCACC CCCAAGAACC 1501 CGTGGAGCAT GGACGAGAAC CTGATGCATA TCAGCTACGA GGCTGGAATC CTGGAGAACC 1561 CCAAGAACCA AGCGCCTCCA GGCCTCTACA CAAAGACCCA GGACCCGGCC AAAGCCCCCA 1621 ACAGCCCGGA CATGCTCGAG ATCGAGTTCA AGAAAGGGGT CCCCGTGAAG GTGACCAACG 1681 TCGGGGATGG CACCACCCAC AGCACAGCGT TGGAGCTTTT CCTGTACCTG AATGAAGTCG 1741 CTGGCAAGCA CGGCGTGGGC CGCATCGACA TCGTGGAAAA CCGCTTCATC GGGATGAAGT 1801 CCCGGGGTAT CTACGAGACC CCAGCGGGGA CGATCCTTTA CCACGCTCAT TTAGACATCG 1861 AGGCCTTCAC CATGGACCGG GAAGTGCGCA AAATCAAGCA AGGCCTCGGC TTGAAATTCG 1921 CCGAGCTGGT CTACACGGGT TTCTGGCACA GCCCCGAGTG TGAATTTGTC CGCCACTGCA 1981 TTGCCAAGTC CCAGGAGCGC GTGGAAGGGA AAGTGCAGGT GTCCGTCTTC AAGGGCCAGG 2041 TGTACATCCT TGGCCGGGAG TCCCCACTGT CCCTCTACAA TGAGGAGCTC GTGAGCATGA 2101 ACGTGCAGGG AGACTACGAG CCGGTTGATG CCACTGGTTT CATCAACATC AATTCCCTCA 2161 GGCTGAAGGA ATATCATCGC CTCCAGAACA AGGTCACCGC CAAAAAGAAT TCTGCAGATA 2221 TCCAGCACAG TGGCGGCCGC TCGAGTCTAG AGGGCCCGCG GTTCGAAGGT AAGCCTATCC 2281 CTAACCCTCT CCTCGGTCTC GATTCTACGC GTACCGGTCA TCATCACCAT CACCATTGAG 2341 TTTAAACCCG CTGATCAGCC TCGACTGTGC CTTCTAGTTG CCAGCCATCT GTTGTTTGCC 2401 CCTCCCCCGT GCCTTCCTTG ACCCTGGAAG GTGCCACTCC CACTGTCCTT TCCTAATAAA 2461 ATGAGGAAAT TGCATCGCAT TGTCTGAGTA GGTGTCATTC TATTCTGGGG GGTGGGGTGG 2521 GGCAGGACAG CAAGGGGGAG GATTGGGAAG ACAATAGCAG GCATGCTGGG GATGCGGTGG 2581 GCTCTATGGC TTCTGAGGCG GAAAGAACCA GCTGGGGCTC TAGGGGGTAT CCCCACGCGC 2641 CCTGTAGCGG CGCATTAAGC GCGGCGGGTG TGGTGGTTAC GCGCAGCGTG ACCGCTACAC 2701 TTGCCAGCGC CCTAGCGCCC GCTCCTTTCG CTTTCTTCCC TTCCTTTCTC GCCACGTTCG 2761 CCGGCTTTCC CCGTCAAGCT CTAAATCGGG GGCTCCCTTT AGGGTTCCGA TTTAGTGCTT 2821 TACGGCACCT CGACCCCAAA AAACTTGATT AGGGTGATGG TTCACGTAGT GGGCCATCGC 2881 CCTGATAGAC GGTTTTTCGC CCTTTGACGT TGGAGTCCAC GTTCTTTAAT AGTGGACTCT 2941 TGTTCCAAAC TGGAACAACA CTCAACCCTA TCTCGGTCTA TTCTTTTGAT TTATAAGGGA 3001 TTTTGCCGAT TTCGGCCTAT TGGTTAAAAA ATGAGCTGAT TTAACAAAAA TTTAACGCGA 3061 ATTAATTCTG TGGAATGTGT GTCAGTTAGG GTGTGGAAAG TCCCCAGGCT CCCCAGCAGG 3121 CAGAAGTATG CAAAGCATGC ATCTCAATTA GTCAGCAACC AGGTGTGGAA AGTCCCCAGG 3181 CTCCCCAGCA GGCAGAAGTA TGCAAAGCAT GCATCTCAAT TAGTCAGCAA CCATAGTCCC 3241 GCCCCTAACT CCGCCCATCC CGCCCCTAAC TCCGCCCAGT TCCGCCCATT CTCCGCCCCA 3301 TGGCTGACTA ATTTTTTTTA TTTATGCAGA GGCCGAGGCC GCCTCTGCCT CTGAGCTATT 3361 CCAGAAGTAG TGAGGAGGCT TTTTTGGAGG CCTAGGCTTT TGCAAAAAGC TCCCGGGAGC 3421 TTGTATATCC ATTTTCGGAT CTGATCAAGA GACAGGATGA GGATCGTTTC GCATGATTGA 3481 ACAAGATGGA TTGCACGCAG GTTCTCCGGC CGCTTGGGTG GAGAGGCTAT TCGGCTATGA 3541 CTGGGCACAA CAGACAATCG GCTGCTCTGA TGCCGCCGTG TTCCGGCTGT CAGCGCAGGG 66

PAGE 85

67 Figure 3 (continued) 3601 GCGCCCGGTT CTTTTTGTCA AGACCGACCT GTCCGGTGCC CTGAATGAAC TGCAGGACGA 3661 GGCAGCGCGG CTATCGTGGC TGGCCACGAC GGGCGTTCCT TGCGCAGCTG TGCTCGACGT 3721 TGTCACTGAA GCGGGAAGGG ACTGGCTGCT ATTGGGCGAA GTGCCGGGGC AGGATCTCCT 3781 GTCATCTCAC CTTGCTCCTG CCGAGAAAGT ATCCATCATG GCTGATGCAA TGCGGCGGCT 3841 GCATACGCTT GATCCGGCTA CCTGCCCATT CGACCACCAA GCGAAACATC GCATCGAGCG 3901 AGCACGTACT CGGATGGAAG CCGGTCTTGT CGATCAGGAT GATCTGGACG AAGAGCATCA 3961 GGGGCTCGCG CCAGCCGAAC TGTTCGCCAG GCTCAAGGCG CGCATGCCCG ACGGCGAGGA 4021 TCTCGTCGTG ACCCATGGCG ATGCCTGCTT GCCGAATATC ATGGTGGAAA ATGGCCGCTT 4081 TTCTGGATTC ATCGACTGTG GCCGGCTGGG TGTGGCGGAC CGCTATCAGG ACATAGCGTT 4141 GGCTACCCGT GATATTGCTG AAGAGCTTGG CGGCGAATGG GCTGACCGCT TCCTCGTGCT 4201 TTACGGTATC GCCGCTCCCG ATTCGCAGCG CATCGCCTTC TATCGCCTTC TTGACGAGTT 4261 CTTCTGAGCG GGACTCTGGG GTTCGCGAAA TGACCGACCA AGCGACGCCC AACCTGCCAT 4321 CACGAGATTT CGATTCCACC GCCGCCTTCT ATGAAAGGTT GGGCTTCGGA ATCGTTTTCC 4381 GGGACGCCGG CTGGATGATC CTCCAGCGCG GGGATCTCAT GCTGGAGTTC TTCGCCCACC 4441 CCAACTTGTT TATTGCAGCT TATAATGGTT ACAAATAAAG CAATAGCATC ACAAATTTCA 4501 CAAATAAAGC ATTTTTTTCA CTGCATTCTA GTTGTGGTTT GTCCAAACTC ATCAATGTAT 4561 CTTATCATGT CTGTATACCG TCGACCTCTA GCTAGAGCTT GGCGTAATCA TGGTCATAGC 4621 TGTTTCCTGT GTGAAATTGT TATCCGCTCA CAATTCCACA CAACATACGA GCCGGAAGCA 4681 TAAAGTGTAA AGCCTGGGGT GCCTAATGAG TGAGCTAACT CACATTAATT GCGTTGCGCT 4741 CACTGCCCGC TTTCCAGTCG GGAAACCTGT CGTGCCAGCT GCATTAATGA ATCGGCCAAC 4801 GCGCGGGGAG AGGCGGTTTG CGTATTGGGC GCTCTTCCGC TTCCTCGCTC ACTGACTCGC 4861 TGCGCTCGGT CGTTCGGCTG CGGCGAGCGG TATCAGCTCA CTCAAAGGCG GTAATACGGT 4921 TATCCACAGA ATCAGGGGAT AACGCAGGAA AGAACATGTG AGCAAAAGGC CAGCAAAAGG 4981 CCAGGAACCG TAAAAAGGCC GCGTTGCTGG CGTTTTTCCA TAGGCTCCGC CCCCCTGACG 5041 AGCATCACAA AAATCGACGC TCAAGTCAGA GGTGGCGAAA CCCGACAGGA CTATAAAGAT 5101 ACCAGGCGTT TCCCCCTGGA AGCTCCCTCG TGCGCTCTCC TGTTCCGACC CTGCCGCTTA 5161 CCGGATACCT GTCCGCCTTT CTCCCTTCGG GAAGCGTGGC GCTTTCTCAT AGCTCACGCT 5221 GTAGGTATCT CAGTTCGGTG TAGGTCGTTC GCTCCAAGCT GGGCTGTGTG CACGAACCCC 5281 CCGTTCAGCC CGACCGCTGC GCCTTATCCG GTAACTATCG TCTTGAGTCC AACCCGGTAA 5341 GACACGACTT ATCGCCACTG GCAGCAGCCA CTGGTAACAG GATTAGCAGA GCGAGGTATG 5401 TAGGCGGTGC TACAGAGTTC TTGAAGTGGT GGCCTAACTA CGGCTACACT AGAAGAACAG 5461 TATTTGGTAT CTGCGCTCTG CTGAAGCCAG TTACCTTCGG AAAAAGAGTT GGTAGCTCTT 5521 GATCCGGCAA ACAAACCACC GCTGGTAGCG GTGGTTTTTT TGTTTGCAAG CAGCAGATTA 5581 CGCGCAGAAA AAAAGGATCT CAAGAAGATC CTTTGATCTT TTCTACGGGG TCTGACGCTC 5641 AGTGGAACGA AAACTCACGT TAAGGGATTT TGGTCATGAG ATTATCAAAA AGGATCTTCA 5701 CCTAGATCCT TTTAAATTAA AAATGAAGTT TTAAATCAAT CTAAAGTATA TATGAGTAAA 5761 CTTGGTCTGA CAGTTACCAA TGCTTAATCA GTGAGGCACC TATCTCAGCG ATCTGTCTAT 5821 TTCGTTCATC CATAGTTGCC TGACTCCCCG TCGTGTAGAT AACTACGATA CGGGAGGGCT 5881 TACCATCTGG CCCCAGTGCT GCAATGATAC CGCGAGACCC ACGCTCACCG GCTCCAGATT 5941 TATCAGCAAT AAACCAGCCA GCCGGAAGGG CCGAGCGCAG AAGTGGTCCT GCAACTTTAT 6001 CCGCCTCCAT CCAGTCTATT AATTGTTGCC GGGAAGCTAG AGTAAGTAGT TCGCCAGTTA 6061 ATAGTTTGCG CAACGTTGTT GCCATTGCTA CAGGCATCGT GGTGTCACGC TCGTCGTTTG 6121 GTATGGCTTC ATTCAGCTCC GGTTCCCAAC GATCAAGGCG AGTTACATGA TCCCCCATGT 6181 TGTGCAAAAA AGCGGTTAGC TCCTTCGGTC CTCCGATCGT TGTCAGAAGT AAGTTGGCCG 6241 CAGTGTTATC ACTCATGGTT ATGGCAGCAC TGCATAATTC TCTTACTGTC ATGCCATCCG 6301 TAAGATGCTT TTCTGTGACT GGTGAGTACT CAACCAAGTC ATTCTGAGAA TAGTGTATGC 6361 GGCGACCGAG TTGCTCTTGC CCGGCGTCAA TACGGGATAA TACCGCGCCA CATAGCAGAA 6421 CTTTAAAAGT GCTCATCATT GGAAAACGTT CTTCGGGGCG AAAACTCTCA AGGATCTTAC 6481 CGCTGTTGAG ATCCAGTTCG ATGTAACCCA CTCGTGCACC CAACTGATCT TCAGCATCTT 6541 TTACTTTCAC CAGCGTTTCT GGGTGAGCAA AAACAGGAAG GCAAAATGCC GCAAAAAAGG 6601 GAATAAGGGC GACACGGAAA TGTTGAATAC TCATACTCTT CCTTTTTCAA TATTATTGAA 6661 GCATTTATCA GGGTTATTGT CTCATGAGCG GATACATATT TGAATGTATT TAGAAAAATA 6721 AACAAATAGG GGTTCCGCGC ACATTTCCCC GAAAAGTGCC ACCTGACGTC AS Overexpression: Experimental plasmids (2 g/well of a 6 well dish) were transiently transfected into BAEC usi ng either Lipofectamine 2000 (Invitrogen) or Fugene (Roche) in serum free Opti-MEM I (In vitrogen) as indicat ed. After 4 h, media

PAGE 86

68 was replaced with Dulbeccos modified Eagl es medium containing 10% serum and cells were cultured for 6 or 24 h. Western Blot: For western blots, equal amounts of protein (measured via BCA assay, Pierce) from clarified lysates were separated on pol yacrylamide gels (BioRad), transferred onto polyvinylidene fluoride memb ranes (PVDF, Millipore) and blotted with indicated antibodies [AS, eNOS, phosphoeNOS (S1177) (BD Biosciences)]. Where appropriate, GAPDH (Novus Bi ologicals) was used as a loading control. RNA Isolation and Real Time PCR: Total RNA was isolated using Tri Reagent following the manufacturers instructions (Sigma). RNA was treated with DNase (Ambion DNA-free). Two g of RNA was reverse transcri bed using the High Capacity cDNA Reverse Transcription Kit per the manufacturers instructions (Applied Biosystems). A 20 l reaction was prepared with 10 l 2X reaction mix (10X RT buffer, dNTPs, 10X RT random primers, 10 X MultiScribe Reverse Transcriptase, RNase inhibitor and nucleas e-free water) and 10 l RNA. The following thermal cycler parameters were utilized for the reverse transcription: 25C for 10 min, 37C for 120 min, 85C for 5 sec and hold at 4C. Real tim e quantitative PCR was performed using the AS specific primers ASL200 and ASR352 utiliz ing a taq-man probe and with the eNOS specific primers eNOSL1075 and eNOSR1226 utilizing SYBR green as described previously by our laboratory [6, 7]. Resu lts were normalized to GAPDH utilizing the primers GAPDHL351 and GAPDHR508. Pr imer sequences are below:

PAGE 87

Table 2: Primers Used for Real Time PCR. ASL200 5-TCA GCA AGG AGT TTG TGG AGG AGT-3 ASR352 5-ACA CAT ACT TGG CTC CTT CTC GCT-3 eNOSL1075 5-TAC ATG AGC ACG GAG ATT GG-3 eNOSR1226 5-AGC ACA GCC AGG TTG ATC TC-3 GAPDHL351 5-CAT GTT TGT GAT GGG CGT GAA CCA-3 GAPDHR508 5-TGA TGG CGT GGA CAG TGG TCA TAA-3 Luciferase Vector Construction: For the luciferase vector, the 189 and 3075 base pair (bp) regions of the AS promoter were constructed as describe d previously [6, 8]. Briefly, luciferase reporter constructs were designed to include the AS promoter and 5UTR up to the AUG start codon cloned upstream of the luciferase gene. The left primers ASL189 (5-GCACTCGAGATCTGCAGGTGGCTGTGAA) and ASL-3075 (5GTACCTCCACTGAAATTGAA) and were combined with ASRluc, (5ATAGAATGGCGCCGGGCG TTTCTTTATGTTTTTGGCGTCTTCCATCGTGACGT GACCAGCGGC) to amplify the AS promoter with an XhoI site on the 5 end and an NcoI site on the 3 end which were used to clone into the vector pGL3Basic (Promega). This strategy took advantage of an NcoI site within the luciferase gene, close to the start codon, to allow for the AS 5-UTR to be cloned adjacent to the start codon. Luciferase Assays: BAEC were cultured as descri bed above and plated in a 24 well plate prior to transfection. Experimental plasmids (200 ng each) and renilla control plasmid pRL-TK (50 ng) were transiently transfected into BAEC using Transit-LT1 (Mirus) in serum free media. After 4 hour s liposomes were removed and incubation continued for 24 hours. Cells were treated with insulin (10 nM, 2 hours). Cells were lysed 69

PAGE 88

70 using passive lysis buffer (Promega). Ten l lysate was assayed for luciferase and renilla activity using a Dual-Luciferase Reporter Assay System (Promega) according to the manufacturers instruct ions. Luciferase and renilla activity were measured as relative light units (RLUs) using a lu minometer (Turner Designs). Al l results were normalized to renilla expression. AS Promoter Analysis: The AS promoter elements that are identified in the discussion were found by doing a promoter sear ch with 2 KB of the AS promoter. The promoter sequence for human AS mR NA with accession number NM_000050 was found utilizing the Transcriptional Regul atory Element Database (TRED) (http://rulai.cshl.edu/cgi-bin/TRED/tred.cgi ?process=home). The promoter sequence was then analyzed utilizing the Transcription Element Search System (TESS) database (http://www.cbil.upenn.e du/cgi-bin/tess/tess). Nitric Oxide Assays: Nitric oxide released into tissue culture medium was measured utilizing the fluorescent probe 2,3-diaminonaphthalene (DAN) as described previously [9]. BAECs were serum starved over night prior to all treatments except in the A23187 experiments. Briefly, at the indicated time points after treatm ent with insulin, A23187 (calcium ionophore), ionomycin (calci um ionophore), bradykinin or VEGF as indicated, aliquots of tissue culture medium were collecte d. Then, freshly prepared DAN reagent in 0.62 M HCl was added to cultu re supernatant, mixed immediately and incubated for 15 minutes. The reaction was st opped with a final concentration of 2.8 M NaOH and the samples were read on a BMG Fluostar Galaxy Spectrofluorometer

PAGE 89

71 exciting at 360 nm and emitting at 405 nm. Total protein was measured via the BCA assay, and data are presented as pm ol nitrite produced per mg protein or as fold change over control. To measure intracellular NO producti on in fixed cells, the NO-specific fluorescen t dye, 4,5-diaminofluor escein diacetate (DAF-2 DA), was utilized as described by Montagnani et al [10] Briefly, BAECs grown to confluence were loaded with DAF-2 DA (final concentration 1 M, 20 min, 37 C) and then rinsed three times, kept in the dark, and maintained at 37 C. Cells were treated as indicated. Fluorescence intensity was im aged utilizing a Nikon Eclipse E1000 Fluorescence Microscope. Statistical Analyses: Statistical analy sis was conducted with a Students T-test of at least 3 independent experiments. Data is pr esented as the average +/the standard error of the mean. Results AS Overexpression Enhances Endothelial NO Production: We hypothesized that since AS is an im portant mediator of the ability of eNOS to produce NO, its overexpression would enhance basal NO producti on. As seen in Figure 4A, transient overexpression of AS utilizing constant amount s of AS plasmid with increasing amounts of transfection reagent led to the genera tion of BAECs that expressed increasing amounts of AS. This led to enhanced producti on of NO that reached a plateau at the mid-

PAGE 90

72 range of AS overexpression, likely due to the inability of eNOS to process the excess substrate without any additional stimulation of its activity. In a similar experiment, the empty vector was also transfected with c onstant plasmid DNA and increasing amounts of transfection reagent (Figure 4B). The results c onfirmed that it is the overexpression of AS per se that leads to the in crease in NO production. These ex periments were carried out after 24 hours of AS overexpr ession and significant cell lo ss was visually noted. To determine whether the cell loss was du e to some condition associated with overexpression or to the enhanced NO produc tion, the experiment was repeated but the NO measurement was conducted after only 6 hours of AS overexpression since no significant cell loss was noted at that time point. As shown in Figure 4C, NO production was enhanced at the 6 hour time point and th is corresponded with a small but detectable overexpression of AS. This suggested that th e cell loss was caused by the AS-mediated increase in NO production. Figure 4D demons trates that AS overexpression does not have a consistent effect eNOS expressi on which suggests the effects on NO production are directly associated w ith the overexpression of AS.

PAGE 91

Figure 4: AS Overexpression Enha nces Nitric Oxide Production. (A) BAEC were serum starved overnight then transfected with 2 g AS plasmid DNA with increasing amounts of Fugene transfection reagent (4, 5, 6, and 8 l which correspond to AS0, AS1, AS2, AS3 and AS4, respectively). Graph demonstrates a nitric oxide assay with data presented as fold NO produced compared to control. Blot demonstrates expression levels of AS that correspond to the NO values. p < 0.007 (n = 3). (B) Nitric oxide assay of BAEC overexpressing AS or an empty vector conducted as in A, except Fugene amounts were 3, 4, 5, 6, 7 and 8 l and this time the empty vector was transfect ed with all conditions. (F3:D2 = 3 l Fugene and 2 g plasmid DNA, etc.). Data is presented as pmol n itrite/mg protein (n = 1). (C) NO assay in cells transiently transfected with the AS expression vector for 6 hours. This time, Lipofectamine 2000 was the transfection reagent used. Graph de monstrates a nitric oxide assay with data presented as fold NO produced as compared to co ntrol. Blot demonstrates expression levels of AS that correspond to the NO va lues (n = 2). (D) Representative blot demonstrating the lack of effect of AS expression on eNOS expression (n = 2). A Overexpressed AS Endogenous AS 0 0.5 1 1.5 2 2.5 3 AS0AS1AS2AS3AS4Fold NO Produced ** 73

PAGE 92

NO Production in Transfected Cells 0 2000 4000 6000 8000 10000 12000 14000 16000 18000 F3:D2F4:D2F5:D2F6:D2F7:D2F8:D2 Fugene to DNA Ratiopmol nitrite/mg protein AS EV B C Overexpressed AS Endogenous AS 0 0.5 1 1.5 2 2.5 CEVWTFold NO Produced 74

PAGE 93

D Overex p ressed AS Endogenous AS eNOS AS Expression and Function are Enhanced by Insulin: Insulin is known to enhance NO production and is a key mediator of vascular health. Part of its atheroprotective properties stem from direct effects on e NOS expression. We hypothesized that AS expression would also be enhanced by insulin. To test this, time course experiments were carried out and it was determined that insulin enhances AS expression most consistently after 2-4 hours of treatment (Figure 5A). We then conducted dose response experiments to test the impact of 1-1000 nM insulin (2 hours) on AS and eNOS expression. Insulin enhanced AS and eN OS expression coordinately. The highest expression was seen at the 100 nM dose, but 10 nM still led to a significant increase in expression (Figure 5 B & C). Additional experiments were conducted with the 2 hour time and 10 nM dose and the effects of insulin on AS expression were confirmed (Figure 5 D & E). 75

PAGE 94

Figure 5: Insulin Increases AS and eNOS Protein Expression. (A) BAEC were serum starved ON and cells were treated with insulin (100 nM) for the indicated times (n = 3). (B) BAEC were serum starved overnight then treated with insulin (1-1000 nM) for 2 hours. Representative western blot demonstrating expression of AS and eNOS. GAPDH was used as a loading control. Data is represented as fold change over control. (C). Densitometry of 2 independent experiments normalized to GAPDH. (D) Representative blot demonstrating expression of AS and eNOS in response to 10 nM insulin treatment for 2 hours. (E) Densitometry of C normalized to GAPDH demonstrating fold change over control (n = 5). p < 0.004 A AS GAPDHControl I 30 min I 1 h I 2 h I 4 h I 6 h AS eNOS GAPDH Insulin B 76

PAGE 95

0 0.5 1 1.5 2 2.5 3 3.5 4 ControlInsulin 1nM Insulin 10nM Insulin 100nM Insulin 1000nMRelative Expression AS eNOS C D AS eNOS GAPDH Control Insulin E 0 0.5 1 1.5 2 2.5 3 AS eNOSRelative Expression Control Ins 10nM * 77

PAGE 96

To determine whether the increase in AS protein expression was related to an increase in mRNA, real time PCR experiments were carried out. The data demonstrated that insulin also increases AS mRNA (Figure 6A). A similar trend was noted for insulin effects on eNOS mRNA. (Figure 6B). Figure 6: Insulin Increases AS and eNOS mRNA Expression. BAEC were serum starved overnight then treated with insulin (10 nM) for 2 hours. (A) Real time PCR results demonstrating AS in control (gray bar) versus insulin treated cells (black bar). p < 0.02 (n = 4). (B) Real time PCR demonstrating increase in eNOS expres sion with insulin treatment. (n = 2) A AS Expression 0 0.5 1 1.5 2 Control Insulin 10 nMRelative Expression eNOS Expression 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 Control Insulin 10 nMRelative Expression B 78

PAGE 97

To determine whether insulin mediates the expression of AS by enhancing activity at specific AS promoter regions, we tested the effect of insulin on a proximal promoter construct (189 bp) versus a full le ngth promoter construc t (3075 bp). Luciferase assay data suggested that there may be distal AS promoter elements regulated by insulin (Figure 7). Figure 7: Insulin Enhances AS Promoter Activity at a Distal Element. BAEC were serm starved overnight then treated with insulin (10 nM) for 2 hours. Luciferase assay reslts comparing 189 versus 3075 base pair promoter constructs. Gray bars represent vehicle treated cells and black bars represent insulin treated cells. Data is presented as relative luciferase units normalized to renilla expression. (n = 1) u u Luciferase Assay 0 1 2 3 4 5 6 189 3075RLU Control Insulin Insulin is known to enhance NO produc tion. To assess whether insulin was leading to an increase in NO production in our system and to correlate increased AS expression to a functional enhancement of th e citrulline-NO cycle, we utilized two methods. First, we treated the cells with insulin, bradykinin, A23187 (calcium ionophore) 79

PAGE 98

80 or a combination of treatments and measur ed NO released into the medium. Although this assay was not sensitive enough to reliably measure the effect of insulin alone, we were able to detect an increase with bradykinin that was en hanced by insulin (Figure 8A). Insulin also demonstrated the ability to enhance NO production above what the calcium ionophore A23187 can do alone (Figure 8B). We then utilized a probe that can measure intracellular NO levels in fixed cells. With this assay, we did see an increase in NO production with insulin treatmen t alone as evidenced by a visually determined increase in green fluorescence (Figure 8C). As expected, th is increase was less than what was seen with the calcium ionophore ionomycin.

PAGE 99

Figure 8: Insulin Enhances Stimulated and Basal NO Production. (A) BAEC were serum starved overnight then treated with insulin (100 nM), bradykinin (10 M) or both for 4 hours. NO assay was conducted by m easuring nitrite released into the medium before and after treatment (n = 3). (B) BAEC were treated with insulin (100 nM), A23187 (1 M) or both for 4 hours. NO assay was conducted as in A (n = 1). (C) BAEC were serum starved overnight then loaded with 1.0 M DAF2-DA for 20 minutes. Cells were then treated with insulin (100 nM) or ionomycin (2 M) for 10 minutes. Cells were fixed and imaged utilizing fluorescent microscopy (ex 480 nM; em 510 nM). A 0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00 ControlINSBKINS + Bk TreatmentFold NO Produced B 0 5 10 15 20 25 30 35 CInsA23187Ins + A23187 TreatmentFold NO Produced Control Insulin Ionomycin C 81

PAGE 100

82 VEGF Regulates AS Expression and E nhances Endothelial NO Production: VEGF is an important mediator of endothelial function. We tested the hypothesis that part of the mechanism of VEGF-mediated vascular protection is associated with an increase in AS expression. First, we measured the time dependency of NO production in response to VEGF. As seen in Figure 9A, af ter 10 minutes of VEGF treatment, there was a statistically significant increase in NO production over untreat ed cells that continued to increase up to our final time point of 2 hours. Then, western bl ot analyses were conducted. The results demonstrated that VEGF increases AS and eNOS expression within 2 hours of treatment (Figure 9B). Importantly, as demonstrated and discussed further in Chapter Three, AS activity is re quired for maximum effects of VEGF on NO production (Figure 16, Page 122).

PAGE 101

Figure 9: VEGF Increases Endothelial Nitric Oxide Production and Upregulates AS and eNOS Expression. (A) Nitric oxide assay measuring NO release into tissue culture medium (pmol nitrite/mg protein) by BAECs treated with VEGF (100 ng/ml) from 5 minutes to 2 hours. Diamonds represent control samples and squares represent VEGF treated samples. p < 0.02; ** p < 0.0003 (n = 3). (B) Representative western blot of BAECs treated with VEGF (100 ng/ml) for 2 hours. Blot demonstrates expression of AS and eNOS. GAPDH was used as a loading control. (n = 2 for AS and n = 1 for eNOS) A 0 1050 2100 3150 4200 5250 6300 7350 8400 9450 10500 05103060120 Timepmol nitrite/mg protein Control VEGF ** ** ** AS eNOS GAPDH Control VEGF B 83

PAGE 102

84 Ceramide Diminishes AS and eNOS Ex pression and Suppresses eNOS Activation: Chronic elevations in ceramide impair vasc ular function and diminish eNOS expression. Often times, ceramide mediates the pathogenic effects of TNF on insulin signaling and other pathways. To test whether chronic ce ramide elevations impact AS and eNOS expression, western blot analyses were conducted. When BAEC were treated with ceramide (10 M) for 16 or 24 hours, AS and eNOS expression was diminished in a time dependent manner (Figure 10A). In addition, the increase in eNOS phosphorylation seen with insulin treatment was diminished when BAEC were pre-treated with ceramide (Figure 10B).

PAGE 103

Figure 10: Ceramide Dimini shes AS and eNOS Expression and Suppresses eNOS Signaling. (A) Representative western blot of BAECs treated with ceramide (10 M) for indicated times after overnight serum starvation. Blot demonstrates expression of AS and eNOS. GAPDH was used as a loading control (n = 1). (B) Representative blot of BAEC that were serum starved for 2.5 hours prior to pre-treatment with ceramide (cer; 10 M) for 5 minutes followed by insulin treatment for 30 minutes (ins; 100 nM). Blot depicts eNOS phosphorylation (peNOS S1177) and total eNOS. GAPDH was used as a loading control (n = 1). AS eNOS GAPDH GAPDH Control Cer 16 h Cer 24 h A B eNOS GAPDH p eNOS Control Ins Ce r Cer + Ins 85

PAGE 104

86 Discussion In this Chapter, we made several impor tant observations. First, we demonstrated that transient overexpression of AS leads to enhanced NO production. This increase was above the basal contribution of the endogenous citrulline-NO cycle. These findings are supported by the work of Xie et. al [11] which demonstrates th at overexpression of AS in vascular smooth muscle cells (VSMC) increases NO production when cells are stimulated with lipopolysaccharid e (LPS) and interferon gamma (IFN ). Under those conditions, the effects they noted were relate d to inducible nitric oxide synthase (iNOS) and not eNOS. Even with stimulation, they saw about a 3-4 fold increase in NO production in stably transfected VSMC. In our case, we were monitoring basal effects and noted ~2 fold increase in NO producti on when AS was transiently overexpressed compared to untransfected or empty vector tr ansfected controls. Our results implied that the effects we saw involved the AS-eNOS axis and not the AS-iNOS axis since a stimulant was not provided to induce iNOS. Th e fact that we saw the effects by 6 hours also suggested that iNOS is not involved si nce such as short time period of time without any stimulation would preclude iNOS induction. Further, the expression of eNOS was not consistently affected by AS overexpression nor did any small variations in eNOS expression correlate with the increases in NO production. This suggests the increase in NO production was a direct result of AS overexpression. Furthermore, our studies support our pr evious work that demonstrates a significant loss of NO production and endothelial cell viability when AS is knocked

PAGE 105

87 down with siRNA [12]. In those studies, redu cing AS expression was directly linked to reduced expression of Bcl-2 and increased caspase activity. Thus, the cell loss was attributed to apoptosis. Further, it was demons trated that apoptosis could be rescued with an NO donor. Therefore, apoptosis was mediat ed by the diminished production of NO as a direct consequence of decreased AS expr ession and activity. The work presented here alludes to the converse scenar io. An increase in AS e xpression led to a significant increase in NO production, despite saturating le vels of NO in the medium and without a change in eNOS activity or expression. In addition, the increase in NO production caused a significant cell loss, likely due to apoptosis. The fact that alterations of AS expression by overexpression or via treatment with various stimulants was associated with enhanced nitric oxide production also s upports the body of evidence that demonstrates that AS is rate limiting for NO production. In addition, it is known that the recycling of arginine from citrulline is the preferred source of substrate for eNOS-mediated NO production, despite available transport systems and excess intracellula r arginine levels [13, 14]. Therefore, our work also supports the moun ting evidence that in endothelial cells, the citrulline-NO cycle is a tightly coupled syst em that generates a dedicated source of arginine for eNOS-media ted NO production [15, 16]. A second set of findings that are collectiv ely important to note is that physiologic (insulin, VEGF) or pathogenic (ceramide) biological molecules affect AS expression coordinately with eNOS. This coordinate regulation is essential for the appropriate level of NO to be produced. Since AS is rate li miting for the production of NO by eNOS [11, 13, 14, 17], it is logical that its expression a nd function must be controlled in a manner

PAGE 106

88 consistent with the substrate needs of eNOS A possible exception to this would be in cases where NO levels have gone beyond physiological concentrations, which might lead to a decrease in AS function in order for NO production to be reduced. Although ultimately, this would lead to a coordinate downregulation of eNOS function, the initial response might be one of discoordinate regulation. Several important biological treatments were explored in this Chapter and each one has important implications. We noted m odest, yet significant increases in AS and eNOS expression in response to insulin. This work is supported by the work of Oyadomari et al. [18] who demonstrated that in ra ts with type 1 diabetes due to streptozotocin (STZ) treatment, AS and e NOS expression in whole aorta is enhanced initially and then decreases with longer dura tion of diabetes. The regulation is coordinate, much like what we see in cell culture mode ls, and suggests the initial increase is a compensatory response while the end result is dysregulation of the citrulline-NO cycle. There is some controversy related to whethe r eNOS expression is diminished in animal models of diabetes or vascular disease [19]. This is likely due to differences in the animal models themselves and the treatment protoc ols utilized. In addition, mechanisms other than expression levels may underlie some fo rms of vascular disorders. Despite the controversy, it is clear that eNOS dysfunction is a hallmark of vascular disorders. Our work certainly demonstrates that AS is also a very important player and more research is needed to delineate the role of AS in diabetic animals and humans.

PAGE 107

89 The role of insulin in regulating AS expre ssion may be due to the direct effects of insulin in regulating the AS promoter. Although the results pr esented regarding enhanced AS promoter activity at a dist al site in response to insu lin are preliminary, there are several AS promoter elements that are known to be insulin responsive such as SP-1, USF and HIF1 [20-22]. Therefore, it is possible that there are several regions in the AS promoter that might regulate the insulin-m ediated upregulation of AS mRNA and protein seen in our studies. In additi on, we saw similar increases in eNOS expression and it has already been demonstrated that eNOS has at least two insulin re sponsive elements: SP-1 and AP-1 [23]. Our work suggests that the functional cons equence of insulin up-regulation of AS expression is an increase in NO production. Th e increases we noted with insulin were generally modest and were most prominent when other stimulants, such as bradykinin, were used. This leads to our hypothesis that insulin is not necessarily a stimulant for NO production. Rather, we believe that the f unction of normal insulin signaling is to allow for NO stimulating pathways to be basally activated and ready for additional stimulation by vasodilators. Without insulin si gnaling, the optimal f unction of stimulants would be blunted. One key deficiency in our understanding of th e upregulation of AS expression by insulin and the concomitant increase in NO production is whether these effects are direct or simply due to known eff ects on eNOS. To fill in this gap, future work is needed to determine whether insu lin increases AS enzymatic activity.

PAGE 108

90 Despite some of the gaps in our understand ing of how insulin directly affects AS activity, our work with VEGF, a nother important mediator of va scular health with direct effects on the endothelium [24], s upports the hypothesis that AS plays a direct role. Like insulin, we found that VEGF leads to a coordi nate upregulation of AS and eNOS protein expression with a concomitant and significan t increase in NO produc tion. Our first hint that AS activity is directly involved comes from experiments that demonstrate that the maximal increase in VEGF-mediated NO production is blunted by a specific AS inhibitor. This data is presented and furthe r discussed in Chapter Three, Figure 16, Page 122. Whether the effect of VEGF on AS prot ein expression is due to an increase on mRNA expression or stability is yet to be de termined. However, there is some evidence in the literature that VEGF regulates the expr ession of some genes such as tissue factor and metallothionein via promoter activation [25]. It is possible there is a similar mechanism in place for AS. Another important finding from this work is the novel link between AS and angiogenic pathways. A lthough the function of AS in arginine regeneration has been associated with tu mor survival and angiogenesis [26], a link between AS activity per se and angiogenic factors has not previously been described. Additionally, the AS-VEGF axis has not been defined in endothelium. Our finding that chronic ceramide treatmen t diminishes AS and eNOS expression demonstrated that pathogenic environments lead to a dysfunction of the citrulline-NO cycle as a whole and not just eNOS. The effects of ceramide on AS expression have not been studied and most studies related to e NOS have determined that ceramide causes inhibition of eNOS activation, not expression. In fact, the only study that assessed eNOS

PAGE 109

91 expression in response to ceramide in huma n endothelial cells found that although NO production was diminished due to elevations in ROS, eNOS expres sion was enhanced in an ineffective compensatory mechanism wh en treated for 10 or 16 hours [27]. In our study, both AS and eNOS expression were dimi nished in a time dependent manner at 16 or 24 hours when using the same dose used by the previous study [ 27]. The discordance of results may be due to the cell type used or the differences in ceramide preparations from different manufacturers. Mo re likely, the issue is related to the fact that ceramide can have both beneficial and i nhibitory effects [28] and this might explain the disparity. Overall, this implies that multiple mechanis ms are disrupted in disease states that ultimately lead to the inabili ty of the endothelium to pr oduce adequate amounts of NO with downstream consequences such as athe rosclerosis or myocardial infarctions. The pathways we have studied related to the regulation of AS expression: TNF PPAR agonists, insulin, VEGF and ceramide are relevant not just individually but also due to the fact that there is a great deal of cross talk between these pathways. One example is that ceramide mediates the pat hogenic effects of chroni c elevations of TNF by inhibiting insulin actions [29]. Also, one mechanism by which PPAR agonists improve insulin sensitivity and improve NO function is by reducing the damaging effects of TNF [30]. VEGF signaling, NO production and angiogenesis are all impaired in insulin resistant conditions [ 31-34]. These associations a nd the pivotal role of AS function in all these pathways improve our global unde rstanding of vascular biology.

PAGE 110

92 Another important observation about AS e xpression is that in tissues where its function is in the urea cycle, such as liver expression is very high [35]. On the other hand, in tissues where AS functions with iNOS to produce NO (such as vascular smooth muscle and glial cells), its expression tends to be very low or even undetectable basally but is highly inducible by cytokines l eading to iNOS inductio n and high output NO production [36-38]. Our findings suggest that in the endothelium, AS expression is basally repressed and this repression is re moved in response to stimuli to increase NO output. This proposed de-repression mechanis m leads to a modest change in AS expression in response to the multiple mechanis ms we have explored up to this point (~ 2-3 fold). It turns out that eNOS transcriptional regulation in general also changes on the order of about 2-3 fold [39]. Although this mi ght seem insignificant, the modest increases in NO associated with this change in e xpression can lead to pronounced changes in vascular tone. Since in the endothelium, eNOS functions to produce constant yet relatively low levels of NO, our model of modest changes in AS expression fits well with this functional paradigm. Thus far, we have found an important pattern of coordinate regulation of AS and eNOS expression. It is important to note in this discussion that the studies led by Laura Pendleton in our lab have uncovered a unique translational regulatory mechanism that involves the expression of differe nt lengths of AS message due to increasing lengths of the 5 UTR [40]. The longer forms of message are endothelium specific and encode for a unique, small protein called Argininosuccinate Synthase Regu latory Protein (ARP). This protein regulates AS translation in trans since it suppresses the tr anslation of the short

PAGE 111

93 form of message that actually encodes for full length AS [7]. So far, this mechanism is unique to AS and seems to allow regulation of tissue-specific AS f unction. Thus, this is one example of regulation of AS that may be completely distinct from eNOS regulation. On the other hand, this may turn out to be a novel global paradigm that allows for the vast complexity of human biology despite a relatively small genome. Overall, the studies presented in this Chap ter further define the central role of AS for NO production and vascular health by uncovering several mechanisms that regulate its expression and lead to concomitant changes in endothelial NO production. These findings are significant since va scular disorders are sometimes characterized by reduced levels of eNOS expression [41-45]. Thus, we predict an important role of AS expression in health and disease. References [1] D. Gospodarowicz, J. Moran, D. Braun, and C. Birdwell, Clonal growth of bovine vascular endothelial cells: fibr oblast growth factor as a surviv al agent, Proc Natl Acad Sci U S A 73 (1976) 4120-4124. [2] H. G. Bock, T. S. Su, W. E. O'Brie n, and A. L. Beaudet, Sequence for human argininosuccinate synthetase cDNA, Nucleic Acids Res 11 (1983) 6505-6512. [3] J. A. Dennis, P. J. Healy, A. L. Beaude t, and W. E. O'Brien, Molecular definition of bovine argininosuccinate s ynthetase deficiency, Proc Natl Acad Sci U S A 86 (1989) 7947-7951. [4] L. C. Surh, A. L. Beaudet, and W. E. O'Brien, Molecular characterization of the murine argininosuccinate s ynthetase locus, Gene 99 (1991) 181-189.

PAGE 112

94 [5] L. C. Surh, S. M. Morris, W. E. O'Brie n, and A. L. Beaudet, Nucleotide sequence of the cDNA encoding the rat argininosuccina te synthetase, Nucleic Acids Res 16 (1988) 9352. [6] B. L. Goodwin, L. C. Pe ndleton, M. M. Levy, L. P. Solomonson, and D. C. Eichler, Tumor necrosis factor-alpha reduces argininosuccinate synthase expression and nitric oxide production in aortic endothelial cel ls, Am J Physiol Heart Circ Physiol 293 (2007) H1115-1121. [7] L. C. Pendleton, B. L. Goodwin, L. P. So lomonson, and D. C. Eichler, Regulation of endothelial argininosuccinate synthase expression a nd NO production by an upstream open reading frame, J Biol Chem 280 (2005) 24252-24260. [8] B. L. Goodwin, K. D. Corbin, L. C. Pendleton, M. M. Levy, L. P. Solomonson, and D. C. Eichler, Troglitazone up-regulates vasc ular endothelial argininosuccinate synthase, Biochem Biophys Res Commun 370 (2008) 254-258. [9] T. P. Misko, R. J. Schilling, D. Salvem ini, W. M. Moore, and M. G. Currie, A fluorometric assay for the measurement of nitr ite in biological samples, Anal Biochem 214 (1993) 11-16. [10] M. Montagnani, H. Chen, V. A. Barr, and M. J. Quon, Insulin-stimulated activation of eNOS is independent of Ca2+ but requi res phosphorylation by Ak t at Ser(1179), J Biol Chem 276 (2001) 30392-30398. [11] L. Xie, and S. S. Gross, Argininosucci nate synthetase overe xpression in vascular smooth muscle cells poten tiates immunostimulant-induced NO production, J Biol Chem 272 (1997) 16624-16630. [12] B. L. Goodwin, L. P. Solomonson, and D. C. Eichler, Argininosuccinate synthase expression is required to maintain nitric ox ide production and cell viability in aortic endothelial cells, J Biol Chem 279 (2004) 18353-18360. [13] L. Xie, Y. Hattori, N. Tume, and S. S. Gross, The preferred source of arginine for high-output nitric oxide synthesis in blood vessels, Semin Perinatol 24 (2000) 42-45. [14] L. J. Shen, K. Beloussow, and W. C. Shen, Accessibility of endothelial and inducible nitric oxide synthase to the intracellula r citrulline-arginine regeneration pathway, Biochem Pharmacol 69 (2005) 97-104. [15] B. R. Flam, D. C. Eichler, and L. P. Solomonson, Endothelial nitric oxide production is tightly coupled to the citrul line-NO cycle, Nitric Oxide 17 (2007) 115-121.

PAGE 113

95 [16] W. C. Sessa, M. Hecker J. A. Mitchell, and J. R. Vane, The metabolism of Larginine and its significance for the biosynthe sis of endothelium-derived relaxing factor: L-glutamine inhibits the gene ration of L-arginine by culture d endothelial cells, Proc Natl Acad Sci USA 87 (1990) 8607-8611. [17] G. Hao, L. Xie, and S. S. Gross, Argininosuccinate synthe tase is reversibly inactivated by S-nitrosylation in vi tro and in vivo, J Biol Chem 279 (2004) 36192-36200. [18] S. Oyadomari, T. Gotoh, K. Aoyagi E. Araki, M. Shichiri, and M. Mori, Coinduction of endothelial nitric oxide syntha se and arginine recycling enzymes in aorta of diabetic rats, Nitric Oxide 5 (2001) 252-260. [19] D. Fulton, M. B. Harris, B. E. Kemp, R. C. Venema, M. B. Marrero, and D. W. Stepp, Insulin resistance does not diminish eNOS expression, phosphorylation, or binding to HSP-90, Am J Physiol Heart Circ Physiol 287 (2004) H2384-2393. [20] T. Kietzmann, A. Samoylenko, U. Ro th, and K. Jungermann, Hypoxia-inducible factor-1 and hypoxia response elements mediat e the induction of plasminogen activator inhibitor-1 gene expression by insulin in primary rat hepatocytes, Blood 101 (2003) 907914. [21] L. M. Cagen, X. Deng, H. G. Wilcox, E. A. Park, R. Raghow, and M. B. Elam, Insulin activates the rat st erol-regulatory-element-bindi ng protein 1c (SREBP-1c) promoter through the combinatorial actions of SREBP, LXR, Sp-1 and NF-Y cis-acting elements, Biochem J 385 (2005) 207-216. [22] M. J. Griffin, and H. S. Sul, Insu lin regulation of fatty acid synthase gene transcription: roles of USF and SREBP-1c, IUBMB Life 56 (2004) 595-600. [23] B. Fisslthaler, T. Benzing, R. Busse, a nd I. Fleming, Insulin enhances the expression of the endothelial nitric oxide synthase in native endothelial cells: a dual role for Akt and AP-1, Nitric Oxide 8 (2003) 253-261. [24] S. Kliche, and J. Waltenberger, VEGF receptor signaling and endothelial function, IUBMB Life 52 (2001) 61-66. [25] B. Joshi, D. Ordonez-Ercan, P. Dasgupta, and S. Chellappan, Induction of human metallothionein 1G promoter by VEGF and heavy metals: differential involvement of E2F and metal transcription factors, Oncogene 24 (2005) 2204-2217. [26] B. J. Dillon, V. G. Prieto, S. A. Curl ey, C. M. Ensor, F. W. Holtsberg, J. S. Bomalaski, and M. A. Clark, Incidence and di stribution of argininosuccinate synthetase deficiency in human cancers: a method for identifying cancers sensitive to arginine deprivation, Cancer 100 (2004) 826-833.

PAGE 114

96 [27] H. Li, P. Junk, A. Huwiler, C. Burkhard t, T. Wallerath, J. Pfeilschifter, and U. Forstermann, Dual effect of ceramide on huma n endothelial cells: induction of oxidative stress and transcriptional upr egulation of endothelial nitric oxide synthase, Circulation 106 (2002) 2250-2256. [28] P. P. Ruvolo, Intracellu lar signal transduction pathwa ys activated by ceramide and its metabolites, Pharmacol Res 47 (2003) 383-392. [29] T. Teruel, R. Hernandez, and M. Lore nzo, Ceramide mediates insulin resistance by tumor necrosis factor-alpha in brown adi pocytes by maintaining Akt in an inactive dephosphorylated state, Diabetes 50 (2001) 2563-2571. [30] S. S. Solomon, L. S. Usdan, and M. R. Palazzolo, Mechanisms involved in tumor necrosis factor-alpha induction of insulin resistance and its reversal by thiazolidinedione(s), Am J Med Sci 322 (2001) 75-78. [31] R. D. Feldman, and G. S. Bierbrier, In sulin-mediated vasodilation: impairment with increased blood pressure and body mass, Lancet 342 (1993) 707-709. [32] M. A. Vincent, M. Montagnani, and M. J. Quon, Molecular and physiologic actions of insulin related to production of nitric ox ide in vascular endothelium, Curr Diab Rep 3 (2003) 279-288. [33] G. Doronzo, I. Russo, L. Mattiello, G. Anfossi, A. Bosia, and M. Trovati, Insulin activates vascular endothelial gr owth factor in vascular smoot h muscle cells: influence of nitric oxide and of insulin re sistance, Eur J Clin Invest 34 (2004) 664-673. [34] F. Bourgoin, H. Bachelard, M. Badeau, S. Melancon, M. Pitre, R. Lariviere, and A. Nadeau, Endothelial and vascular dysfunctions and insulin resistance in rats fed a highfat, high-sucrose diet, Am J Physiol Heart Circ Physiol 295 (2008) H1044-H1055. [35] A. Husson, C. Brasse-Lagnel, A. Fairand, S. Renouf, and A. Lavoinne, Argininosuccinate synthetase from the ur ea cycle to the citrulline-NO cycle, Eur J Biochem 270 (2003) 1887-1899. [36] Y. Hattori, E. B. Campbell, and S. S. Gross, Argininosuccinate synthetase mRNA and activity are induced by immunostimulants in vascular smooth muscle. Role in the regeneration or arginine for nitric oxide synthesis, J Biol Chem 269 (1994) 9405-9408. [37] A. K. Nussler, T. R. Billiar, Z. Z. Liu, and S. M. Morris, Jr., Coinduction of nitric oxide synthase and argininos uccinate synthetase in a murine macrophage cell line. Implications for regulation of nitric oxide production, Journal of Biological Chemistry 269 (1994) 1257-1261.

PAGE 115

97 [38] A. Schmidlin, and H. Wiesinger, Argi ninosuccinate synthetase: localization in astrocytes and role in the producti on of glial nitric oxide, Glia 24 (1998) 428-436. [39] C. D. Searles, Transcri ptional and posttr anscriptional regulation of endothelial nitric oxide synthase expression, Am J Physiol Cell Physiol 291 (2006) C803-816. [40] L. C. Pendleton, B. L. Goodwin, B. R. Flam, L. P. Solomonson, and D. C. Eichler, Endothelial argininosuccina te synthase mRNA 5'-untra nslated region diversity. Infrastructure for tissue-speci fic expression, J Biol Chem 277 (2002) 25363-25369. [41] N. Makino, T. Maeda, M. Sugano, S. Sa toh, R. Watanabe, and N. Abe, High serum TNF-alpha level in Type 2 diabetic patient s with microangiopathy is associated with eNOS down-regulation and apoptosis in endothelial cells, J Diabetes Complications 19 (2005) 347-355. [42] L. Agnoletti, S. Curello, T. Bachet ti, F. Malacarne, G. Gaia, L. Comini, M. Volterrani, P. Bonetti, G. Parrinello, M. Cadei, P. G. Grigolato, and R. Ferrari, Serum from patients with severe heart failure downre gulates eNOS and is proapoptotic: role of tumor necrosis factor -alpha, Circulation 100 (1999) 1983-1991. [43] M. Tsutsui, H. Shimokawa, T. Morish ita, Y. Nakashima, and N. Yanagihara, Development of genetically engineered mice lacking all three nitric oxide synthases, J Pharmacol Sci 102 (2006) 147-154. [44] B. S. Oemar, M. R. Tschudi, N. Godoy, V. Brovkovich, T. Malinski, and T. F. Luscher, Reduced endothelial nitric oxide synthase expression a nd production in human atherosclerosis, Circulation 97 (1998) 2494-2498. [45] T. Fujii, M. Onimaru, Y. Yonemitsu, H. Kuwano, and K. Sueishi, Statins restore ischemic limb blood flow in diabetic micr oangiopathy via eNOS/NO upregulation but not via PDGF-BB expression, Am J Ph ysiol Heart Circ Physiol 294 (2008) H2785-2791.

PAGE 116

98 CHAPTER TWO ARGININOSUCCINATE SYNTHASE PHOSPHORYLATION Overview Argininosuccinate synthase (AS) is essential for endot helial nitric oxide (NO) production and its regulation in this capaci ty has been studied primarily at the transcriptional level. The dynamics of vascular function suggest that an acute regulation system may mediate AS function. This pr emise underlies our hypothesis that AS is phosphorylated in vascular endothelium. Sin ce serine/threonine phosphorylation has been identified as the key mechanism mediating acute nitric oxide pr oduction, we focused on these modifications. We began our studies by conducting a bioinformatic analysis of the AS protein sequence utilizing 4 different data bases. We identified 31 putative sites of AS serine/threonine phosphorylation th at were positive hits in at least 2 of the 4 databases. Immunoprecipitation and immobilized metal affi nity chromatography demonstrated that AS is an endogenous phosphoprotein. An in vitro kinase screen reve aled that protein kinase A (PKA) and prot ein kinase c alpha (PKC ), kinases that enhance NO production via eNOS activation, phosphorylate AS. Vascul ar endothelial growth factor (VEGF) was identified as a candidate pathway for regul ating AS phosphorylation since it activated PKA, PKC and eNOS. In addition, -methyl-DL-aspartic acid (MDLA), an AS

PAGE 117

99 inhibitor, diminished maximal VEGF-m ediated NO production. Immunoprecipitation studies suggested that VEGF enhances AS phosphorylation. We then focused our studies on identifying specific sites of AS phosphorylation utilizing proteomics. Thus far, we have been able to identify the following sites of AS serine/threonine phosphorylation: T131, S134, S180, S189, S328. Analysis of the AS 3-dimensional structure revealed important structure-function hypot heses related to the identified sites. Site directed mutagenesis of T131, S180 and S189 to gene rate phospho-null (alanine substitution) and phospho-mimetic (aspartic acid substitution) va riants revealed a decrease and partial recovery of NO production compared to wild type, respectively. In silico modeling of phosphorylation sites utilizing the human AS crystal struct ure revealed that T131, S134 and S328 are the most accessible sites for m odification by phosphorylation. Overall, our data demonstrates that regulation of AS by serine/threonine phosphorylation is an important mechanism for ensuring NO homeostasi s. Our work suggests that targeting the kinases that lead to AS phosphorylation is one avenue to normalize AS function and offers an attractive therapeutic option for vascul ar disorders. Materials and Methods Bioinformatics: The following bioinformatic databa ses were utilized to identify putative AS phosphorylation site s and kinases: NetworKIN (http://networkin.info/search.php), Phospho-Motif (http://www.hprd.org/PhosphoMotif_finder), NetPhosK (http://www.cbs.dtu.dk/services/NetPhosK/ ) and Group Based Prediction System

PAGE 118

100 (http://bioinformatics.lcd-ustc.org/gps2/). Addi tionally, kinase motifs were identified by searching for specific motifs in the AS seque nce utilizing Expasy Prosite (Scan Prosite Tool: http://www.expasy.ch/tools/scanprosit e/). Multiple sequence alignments were conducted utilizing ClustalW (http://www.ebi.ac.uk/Tools/clustalw2/index.html). The 3dimensional human AS crystal structure was generated utilizing Vi ewerLite software version 5.0 (Accelrys Corporation, San Diego, California) or iSee software from the Structural Genomics Consortium (http://www.sgc.ox.ac.uk/iSee/). Bovine Aortic Endothelial Cell Culture: See Chapter One, Page 64 In Vivo 32P Orthophosphate Labeling: BAEC were serum starved for 4 hours in phosphate-free DMEM. Cells were transfected with V5-His tagged AS for 24 hours (see Chapter One, Page 64 for a full description of this vector) and then biosynthetically labeled with [32Pi] according to the procedure of Michel et al [1]. Labeled cells were treated with 10 M bradykinin for 1 hour. Cell ly sates were prepared and immunoprecipitations were carried out using an tibodies directed against the V5-tag of overexpressed AS. Following the isolation of immune complexes with protein G-agarose (Santa Cruz), the complex was eluted by heating in SDS-PAGE sample buffer. The samples were fractionated via SDS-PAGE (420% gradient gels; Bio-Rad). After drying the gel, detection of phosphorylated AS was via Phosphor Imaging (GE Healthcare). Immunoprecipitation and Western Blot: To determine whether AS is an endogenous phosphoprotein, immunoprecipitations (IP) were carried out utilizing a

PAGE 119

101 phospho-serine/threonine an tibody (BD Biosciences) and an AS antibody (Everest Biotech). To assess regulation of AS phos phorylation by VEGF, IP were carried out utilizing an AS antibody (Everest Biotech). After serum starvation overnight, complete medium with serum was replenished for 2 hours to restore maximal basal phosphorylation levels or cells were treated with VEGF (100 ng/ml) as indicated in figure legends. BAEC lysates were prepared from 150 mm dishes by addi ng 1 ml NP-40 buffer (50 mM Tris HCl, pH 7.4, 150 mM NaCl, 1% NP-40) plus protease and phosphatase inhibitors (Calbiochem and Pierce, respectivel y) and scraping. Lysates were centrifuged at 12,000 x g at 4C for 10 minutes. Equal am ount of total protein (measured by BCA assay, Pierce) from clarified lysates was incubated with 200 l Dynabeads protein G (Invitrogen) per ml lysate at 4C for 1 hour for pre-clearing. Supernat ants were collected and incubated with correspondi ng antibodies at 4C for 4 hours. The antibody-antigen complex was then incubated with 200 l/ml Dynabeads Protein G for 1 hour at 4C. IPs with same species normal IgG antibodies (S anta Cruz) were conduc ted in parallel as negative controls. Whole cell lysate (WCL) was utilized as a positive control for antibody reactivity. For western blotting methods, refer to Chapter One, Page 68. Membranes were blotted with indicated antibodies [eNOS phospho-eNOS (S1177) (BD Biosciences); PKA, phospho-PKA (T497), phospho-PKA subs trate, phospho-PKC substrate, phosphoserine/threonine-phenylalanine (Cell Signaling Technologies); phospho-serine (Millipore)]. Where appropriate, GAPDH (Novus Biologicals ) was used as a loading control.

PAGE 120

102 Affinity Chromatography: Two different methods were utilized to separate phosphorylated and non-phosphorylated protein. First, the Phosphoprotein Enrichment Kit from Pierce was utilized. Procedures were conducted according to supplied instructions. In brief, cells were washed in HEPES buffered saline and lysed in included lysis buffer + 0.25% CHAPS and protease/phos phatase inhibitors (HALT Protease and Phosphatase inhibitor cocktails, Pierce). An equal amount of protein (measured with the Coomassie Plus Protein Assay; Pierce) di luted to 0.5 mg/ml was placed in supplied columns and incubated on a platform rocker for 30 minutes at 4C. The flow through was collected and after washing, phos phoproteins were eluted w ith supplied elution buffer. Flow through and eluted fractions were conc entrated using iCON C oncentrators (Pierce). Total protein was measured and western blot s were conducted as above. The presence of AS in the phosphoprotein fraction was monitored using a total AS antibody (BD Biosciences). To demonstrate effectiven ess of separation and enrichment of phosphoproteins, membranes were probed w ith [phospho-Akt (S473) (Cell Signaling Technologies)]. Additionally, cytochrome C (Cell Signaling Technologies), a nonphosphorylated protein, was used as a negative control. For the -phosphatase experiments, once cell lysates were collected, cells were treated 700 g of -phosphatase per 100 g protein for 20 minutes at 37C. The IM AC procedure then was continued as above. The second method involved the Qiagen PhosphoProtein Purification Kit. Procedures were according to supplied dire ctions. In brief, after collecting cells by trypsinization and washing cell pellet in HE PES buffered saline, cellular proteins were

PAGE 121

103 extracted in lysis buffer cont aining 0.25% (w/v) CHAPS, prot ease/phosphatase inhibitors, and benzonase for 30 min at 4C and centrifuged at 10,000 g at 4C for 30 min to remove insoluble material. Total protein was diluted to a concentration of 0.1 mg/ml in a total of 25 ml of lysis buffer and was applied to a lysis buffer-equilibrated PhosphoProtein purification column at room temperature. Th e column was washed with lysis buffer and the phosphoproteins were eluted with 2 ml of PhosphoProtein Elution Buffer. The yield of phosphorylated protein was determined by the BCA assay. The flow-through samples were passed through two additional column s to ensure complete removal of phosphoproteins from the sample. The eluted and flow-through fractions were then concentrated by ultrafiltrati on in a 10-kDa cutoff Amicon Ultra column (Millipore Corporation) and equal amount of protein from each sample was subjected to SDSPAGE. The presence of AS in the phosphoprotein fraction was monitored using a total AS antibody (BD Biosciences). To demonstr ate effectiveness of separation and enrichment of phosphoproteins, membranes were probed with phospho-eNOS (S1177) (BD Biosciences). -tubulin was utilized as a loading control. Purification of Bovine Argininosuccinate Synthase: Bovine AS (NP_776317) was subcloned into the pET-28(c)+ v ector (Novagen) and expressed in E. coli. The protein was subsequently purified via the fused His tag by affinity chromatography utilizing Ni-NTA His-bind resin per the manufacturers instructions (Novagen). Successful purification was veri fied via SDS-PAGE. Purity was ~90% as determined by densitometry as shown below:

PAGE 122

Figure 11: SDS-PAGE Demonstrating AS purification AS In Vitro Kinase Screen: Purified AS was utilized to run in vitro kinase reactions according to the manufacturers instructions (SignalChem, Vancouver, BC). Kinases were selected by conducting a bioinformatic search for putative AS kinases and then choosing a subset of those kinases for the sc reen based on their known role in regulating NO production. The following kinases were screened: AMP-activated protein kinase (AMPK; subunits A1/B1/G1 and A1/B1/G2 ), casein kinase II (CKII; subtype 1), glycogen synthase ki nase 3 beta (GSK3 ), protein kinase A (PKAcatalytic unit c ), protein kinase C alpha (PKC ) and protein kinase G (PKGsubtype 1). In brief, reactions contained the following components: active pr otein kinase, 10X reaction buffer, protein kinase activator, 33P-ATP and 5-10 g recombinant, His-tag purified AS. A positive control to test for kinase activity was carried out with the appropriate peptide substrate and a blank reaction was carri ed out with all assay com ponents except substrate. The assays were initiated by the addition of 33P-ATP and the reaction mixtures were incubated at 30C for 45 minutes. The assays were terminated by spotting the reaction mixture onto a phosphocellulose P81 plate. After 3 15-minute washes in 1% phosphoric 104

PAGE 123

105 acid, the radioactivity on the P81 plates was counted via scintillation counting. For imaging, in vitro kinase reactions were ca rried out essentia lly as above except they were terminated upon addition of SDS-PAGE sample buffer. Reactions were fractionated on 10% SDS-PAGE gels (BioRad) and exposed to film. For the dose response experiments, the reactions were carried out as above, w ith the exception that AS protein (substrate) concentration was varied from 2.5-12.5 g. To determine if pre-phosphorylation of AS was required for CKII or GSK3 to phosphorylate AS, an in vitro phosphorylation reaction was carried out with AS and PKC using cold ATP followed by the in vitro reaction with CKII or GSK3 utilizing radiolabeled ATP. Nitric Oxide Assays: See Chapter 1, Page 70 for the methodology employed to measure NO released into the medium utilizing the DAN assay. For the VEGF experiments, cells were treated w ith VEGF +/an AS inhibitor, -methyl-DL-aspartic acid (MDLA; Sigma) as indicated. The experime nts with wild type and mutant AS were carried out by overexpressing AS (see below). Generation of AS Variants and Transient Transfections: The AS expression vector, fully described in Chapter One, Pa ge 64, was utilized to mutate identified phosphorylation sites to determ ine their role in NO produc tion utilizing procedures routinely conducted by our laboratory [2]. Briefly, phosphorylated residues were mutated to alanines to mimic a non-phosphorylated state and to aspartic acid to mimic a constitutively phosphorylated st ate following the Quick Cha nge protocol utilizing Pfu Turbo DNA Polymerase (Stratagene). We co mpared the NO production of wild type

PAGE 124

versus phospho-mutant AS in BAECs since we have demonstrated a consistent and reproducible increase in NO production when w ild type AS is overexpressed in BAECs (See Chapter One, Figure 4, Page 73). Primers were designed according to th e guidelines from Strategene for the successful use of their Quick Change protocol. For the S180A/S189A, the S189A variant was used as the template and the S180A primers were used to introduce the additional mutation. All variants were veri fied by sequencing. The following primers were utilized: Table 3: Primers Used for Site Directed Mutagenesis Variant Wild Type Sequence Sense Primer Antisense Primer T131A C CGG TTT GAG CTC GCC TGC TAC TCG CTG G C CAG CGA GTA GCA GGC GAG CTC AAA CCG G T131D C CGG TTT GAG CTC ACC TGC TAC TCG CTG G C CGG TTT GAG CTC GAC TGC TAC TCG CTG G C CAG CGA GTA GCA GTC GAG CTC AAA CCG G S180A CAA GAA CCC GTG G GC C AT GGA CGA GAA CCT G C AGG TTC TCG TCC ATG GCC CAC GGG TTC TTG S180D CAA GAA CCC GTG G AG CAT GGA CGA GAA CCT G CAA GAA CCC GTG G GA C AT GGA CGA GAA CCT G C AGG TTC TCG TCC ATG TCC CAC GGG TTC TTG S189A AGA ACC TGA TGC ATA TCG CCT ACG AGG CTG GAA TCC AGA ACC TGA TGC ATA TC G ACT ACG AGG CTG GAA TCC S189D AGA ACC TGA TGC ATA TC A GCT ACG AGG CTG GAA TCC GGA TTC CAG CCT CGT A GG CGA TAT GCA TCA GGT TCT GGA TTC CAG CCT CGT AGT CGA TAT GCA TCA GGT TCT Experimental plasmids were transi ently transfected into BAEC using Lipofectamine 2000 (Invitrogen) in seru m free Opti-MEM I (Invitrogen). Two g AS plasmid DNA was used per we ll of a 6-well dish or 10 g per 100 mm dish. After 4 h, the 106

PAGE 125

107 medium was replaced with Dulbeccos modi fied Eagles medium containing 10% serum and cells were cultured for 24 h. Purification of Overexpressed AS: Two methods were utilized to partially purify the AS expression vector. In order to obtai n sufficient material for mass spectrometry analysis, at least 3 100-mm dishes were tr ansfected per treatment group as described above and lysates were pooled. Cells were tr eated with either 20 or 100 nM okadaic acid, a serine/threonine phosphatase i nhibitor, in an attempt to enhance phosphorylation signal for the experiments to characte rize the AS expression vector. The first method involved immunoprecipita tion of overexpressed AS with a V5 antibody. Lysates were prepared utilizing NP40 lysis buffer (50 mM Tris HCl, pH 7.4, 150 mM NaCl, 1% NP-40) plus protease and phosphatase inhibito rs (Calbiochem and Pierce, respectively) and scraping. Clarified lysates were pre-cleared with a protein G bead slurry (Protein G Plus Agarose, Santa Cruz) for 1 hour. Immunoprecipitation was conducted with an antibody against the V5 tag of the AS expression construct (Invitrogen) by incubating the pre-cleared lysate with antibody for 2 hours at 4C. The protein G bead slurry was added to samples a nd incubated overnight at 4C. Purified AS and associated proteins were eluted with 2X SDS-PAGE sample buffer. The second method involved purification via the 6X-his tag of the AS expression vector. Magnetic Ni-NTA agarose beads (Qiagen) were utilized according to the manufacturers instru ctions. Briefly, cells were lyse d in Buffer B-Tween + 1% NP-40

PAGE 126

108 (100 mM NaH2PO4, 10 mM TrisCl, 8 M urea and 0.05% tween, pH 8.0). Lysates were cleared by centrifugation. Ni-NTA agarose beads were added (20 l/ml) to the lysates and incubated with end-over-end rotation for 2 hours at 4C. Cells were washed in Buffer C-Tween (same as buffer B, pH 6.3) and eluted with Buffer E-Tween (same as buffer B, pH 4.5). To monitor expression levels, wester n blots were conducted as described previously (purified samples and whole cell lysates were compared) and membranes were probed with an AS antibody. To assess whethe r the purified vector was phosphorylated, membranes were probed with a phospho-serine antibody (Zymed). To visualize effectiveness of purification, duplicate gels were run, coomas sie stained, dried by vacuum and photographed. Liquid Chromatography and Tandem Mass Spectrometry: AS was transiently overexpressed in BAECs. Twenty four hours af ter transfection, cells were deprived of serum overnight. Cells were treated with in sulin, bradykinin, insu lin + bradykinin and okadaic acid. Overexpressed AS was purifie d by its fused 6X-His tag using Ni-NTA agarose magnetic beads (Qiagen) as describe d above. Proteins we re separated by SDSPAGE to identify the AS band (51 kD). A duplicate gel was run to confirm expression of the AS plasmid by western blot Gel bands of interest were excised and destained. The protein disufides were reduced with tris carboxyethylphosphine and then the cysteines were alkylated with iodoacetamide. In-gel tr ypsin digestion was used for proteolysis. The resulting peptides were extracted and concentrated pr ior to liquid chromatography

PAGE 127

109 coupled to tandem mass spectrometry (LC-MS /MS) analysis. Nanoflow reverse phase liquid chromatography was used to sepa rate the peptides by hydrophobicity (LC Packings, Dionex, Sunnyvale, CA). Online detection was accomplished with an electrospray linear ion trap mass spectrometer (LTQ, Thermo, San Jose, CA). Peptide molecular weight measurements preceded i on selection, fragmenta tion, and fragment ion detection in MS/MS. Tandem mass spectra were assigned to peptide sequences using Mascot and Sequest database search algorithms. Sequence assignments were validated by manual inspection of the data. In-Silico Modeling of AS Three Dimensional Structure: In silico models of AS with identified phosphorylation si tes were generated by the Un iversity of South Florida Department of Chemistry by Dr. Wayne Guida and Daniel Santiago utilizing a Molecular Dynamics approach. Modeling was dependent on the original X-ray structure (PDB ID: 2NZ2). Substrate binding affinity was measured when sites were not phosphorylated versus phosphorylated (comparative docking) The methodology is composed of four phases: I. Substrates were docked to original enzyme structure. II. Phosphorylation models were created fo r each possible serine /threonine residue. III. Low energy conformer for each phosphorylation model was identified. IV. Substrates were docked into phosphorylation models and docking scores were compared. Statistical Analyses : See Chapter One, Page 71.

PAGE 128

110 Results AS is an Endogenous Phosphoprotein: Since serine/threonine phosphorylation is a prominent mechanism that regulates acute eNOS function, we focused on these modifications. To assess whether endogenous AS is phosphorylated, several approaches were taken. First, a bioinformatic sear ch for AS serine/threonine phosphorylation utilizing four different databases was c onducted. The benefit of utilizing multiple databases is that they each u tilize slightly different approaches. For example, NetPhos 2.0 utilizes trained neural netw orks for predictions [3]. On the other hand, NetworKin predicts in vivo kinase-substrate relationships by a ugmenting the information gained from kinase substrate motifs with context for kinases and phosphoproteins [4]. Phospho-Motif predictions are based on known consensus kina se motifs as well as additional motifs that are curated from the literature [5]. Finally, Group Based Prediction System improves on the methods of standard prediction syst ems by grouping kinases into hierarchical structures and developing and approach to minimize false positives [6]. By cross referencing results from the 4 databases, the number of false positives should be reduced. Through these in silico experiments, 31 sites were identified that were positive hits in at least 2 of the 4 databases (Table 4).

PAGE 129

T able 4: Predicted AS serine/threonine phosphorylation sites. Data was generated by ross-referencing results from 4 different databases and choosing only sites that were positive hits in at least 2 of 4 databases. positive hits in at least 2 of 4 databases. c 111 Position Position Peptide Peptide 2 --MS GKG 6 GKGS VVL 17 GLDT SCI 65 EDIS KEF 80 IQSS ALY 91 LLGT SLA 92 LGTS LAR 115 KYVS HGA 119 HGAT GKG 131 FELS CYS 174 VPVT PKN Position Peptide 180 NPWS MDE 189 MHIS YEA 208 GLYT KTQ 210 YTKT QDP 219 APNS PDM 242 GDGT THS 243 DGTT HST 246 THST ALE 278 GMKS RGI 284 IYET PAG 288 PAGT ILY Position Peptide 301 EAFT MDR 328 FWHS PEC 341 IAKS QER 352 VQVS VFK 365 GRES PLS 368 SPLS LYN 376 ELVS MNV 396 NINS LRL 410 NKVT AKTo evaluate whether the in silico identification of AS phosphorylation was occurring endogenously, immunopr ecipitation of BAEC lysate s was carried out with a phospho-serine/threonine antibody followed by immunoblotting with an AS antibody. As seen in Figure 12, AS co-immunoprecipitat es when cellular proteins that are phosphorylated at serine and/or threonine residues are enriched. Similar results were obtained with the reverse immunoprecipitation, but in this case, the best results were obtained when probing with a phospho-serine and a phospho-serine/threonine/tyrosine antibody (Figure 12). In addition, in vivo 32P orthophosphate labeling was utilized to determine if AS is phosphorylated. AS was overexpressed, labeled with 32P then treated with bradykinin. Lysates were prepared and overexpressed AS was purified via immunoprecipitation. As shown in Figure 12, AS is phosphorylated basally and this increased upon bradykinin treatment.

PAGE 130

Figure 12: AS is an Endogenous Phosphoprotein, Part I. (A) Representative blot (n = 3) of immunoprecipitation of BAEC lysates utiliz ing a phospho-serine/threonine antibody (IP pS/T) or a same-species normal IgG (IP IgG) The membrane was probed with an AS antibody (IB AS). WCL; whole cell lysate. (B) Representative blot (n = 3) of immunoprecipitation of BAEC lysates utilizing an AS antibody (IP AS) or a same-species normal IgG (IP IgG). The membrane was probed with a phospho-serine or a phosphoserine/threonine/tyrosine antibody (IB pS and IB pS/T/Y, respectively); pAS = phospho-AS. (C) Phosphor-image demonstrating the level of 32P incorporation in response to bradykinin stimulation (10M) in BAECs where either AS (AS-V5-His) or an empty vector were overexpressed and immunoprecipitated (n = 1). B IB pS/T/Y IB pS IPAS IP IgG IgG pAS pAS A IB AS IP pS/T IP IgG WCL C AS-V5-His Empty Vector Control Bradykinin pAS To further validate that AS is ind eed an endogenous phosphoprotein in the endothelium, we utilized immobilized metal a ffinity chromatography (IMAC) to separate and enrich phosphorylated protein from total protein. This was done with two different 112

PAGE 131

113 methods since the different resi ns utilized to bind phosphorylat ed proteins have affinity for some phosphoproteins and not others [7]. With the first method, the western blot analysis in Figure 13A demonstrated that a fraction of AS eluted from the column as a phosphorylated protein. As a control, membranes were re-probed with a phospho-Akt antibody. Phospho-Akt was observed in the elut ed phosphoprotein fraction with a relative absence in the flow-through fraction (Figure 13A). As a negative control, membranes were re-probed with a cytochrome c anti body (cytochrome c is not phosphorylated) and no cytochrome c was observed in the elut ed phosphoprotein fraction (Figure 13A). Figure 13B demonstrates the similar results that were obtai ned with the second method. A fraction of AS eluted as a phospho-protein. This time, phospho-eNOS was utilized as a positive control to demonstrat e the effectiveness of the separation of phosphoprotein from total protein (Figure 13B ). In addition, these experiments were conducted with various treatments in an attempt to monitor changes in AS phosphorylation. There were no obvious changes in the amount of AS eluted in the phosphoprotein fraction. Because of the high affinity of these resins for phosphorylated proteins, it is nearly impossi ble to detect changes in bindi ng to the column with a total protein antibody since a treatment may cause a change in phosphoryl ation sites rather than a net increase or decrease in phosphorylatio n. It became evident that this technique could not be used to monitor changes in phosphorylated AS without having a phosphoAS antibody. This was demonstrated in subs equent experiments where dephosphorylation reactions were ca rried out with -phosphatase prior to the IMAC experiments. While decreases in phosphorylation were noted wh en using a site specific phospho-Akt

PAGE 132

114 antibody, there was little to no change in th e amount of AS eluted when a total AS antibody was used. In order for these experiments to have worked, AS needed to be completely dephosphorylated prior to IMAC in order to see a significant change. Finally, to make sure that the IMAC data was not an artifact, IMAC was conducted and then AS was immunoprecipitate d from the eluted phosphoprotein fraction. As shown in Figure 13C, AS was successfu lly immunoprecipitated from the eluted fraction suggesting that the signal noted with IMAC is indeed a fraction of AS that elutes as a phosphoprotein and not simply a non-spec ific band at the AS molecular weight. Taken together, the immunoprecipitation, 32P labeling and phosphoprotein affinity chromatography strongly suggest that AS is an endogenous phosphoprotein in vascular endothelium.

PAGE 133

Figure 13: AS is an Endogenous Phosphoprotein, Part II. (A) Representative blot (n = 3) demonstrating separation of phosphorylated protein from total protein utilizing immobilized metal affinity chromatography (IMAC). The top panel shows the membrane was probed with AS. Phospho-Akt was utilized as a positive control and cytochrome c was used as a negative control. (B) Representative blot of IMAC conducted as in (A) but with a different resin (n = 3). The blot shows the level of AS, phospho-eNOS or -tubulin in the eluted (E) or flow through (FT) fractions in response to treatment with insulin (100 nM), okadaic acid (50 nM; OA) or H89 (n = 3). (C) Representative blot of the eluted phospho protein fraction of BAECs treated without (vehicle) or with -phosphatase ( -pptase) prior to IMAC conducted as in (A) (n = 2). (D) Representative blot demonstrating IMAC conducted as in (B). The data represents AS expression in the whole cell lysate (WCL), the eluted IMAC fraction (IMAC) or from the IMAC fraction after immunoprecipitation with an AS antibody (IMAC + IP) in cells treated with okadaic acid (OA) at the doses indicated (n = 1). A AS pAkt Cyt c Phosphoprotein Non-Phosphoprotein Total Protein AS peNOS -tubulin Control, E Control, FT Insulin, E Insulin, FT OA, E OA, FT H89, E H89, FT B AS pAkt Vehicle -pptase C 115

PAGE 134

IMAC + IP WCL IMAC AS Control OA 20 nM OA 100 nM Control OA 20 nM OA 100 nM Control OA 20 nM OA 100 nM D Biological Relevance of AS Phosphorylation: In order to establish the biological significance of AS phosphoryla tion, we conducted a targeted in vitro kinase screen. First, we identified putative AS kinase s utilizing four bioinformatic databases. From that list of kinases, the following were selected for the sc reen due to their known role in nitric oxide biology: AMP-activated protein kinase (A MPK), casein kinase II (CKII), glycogen synthase kinase 3 beta (GSK3 ), protein kinase A (PKA) and protein kinase C alpha (PKC ). Akt1 was used as a negative control due to the lack of the consensus Akt phosphorylation motif in the AS protein seque nce (R-X-R-X-X-S/T). The kinase screens were carried out with purified recombin ant bovine AS and as shown in Figure 14A, PKC and PKA were identified as AS kinases due to their ability to phosphorylate AS in vitro Although there was an increase in count s above the blank reac tion with AMPK and Akt1, this was below the threshold considered significant (Figure 14A). SDS-PAGE fractionation of in vitro kinase reaction products and visualization by film exposure confirmed that PKC and PKA, but not Akt1, phosphorylate AS (Figure 14B). Repeat in vitro kinase reactions for GSK3 and CKII were carried out, but AS was pre116

PAGE 135

phosphorylated with PKC CKII and GSK still were unable to phosphorylate AS in vitro (Figure 14C). Figure 14: AS is Phosphorylated by PKC and PKA. (A) In vitro kinase assays. Gray bars represent the incubation of purified AS minus kinase (Blank). Black bars represent the incubation of purified AS with the indicated kinase (Kinas e). Kinase activity is represented as counts per minute (CPM). p < 0.0002; ** p < 0.00003; (n = 3 for PKC, PKA and Akt; n = 1 or 2 for remaining kinases). (B) Representative image of in vitro kinase reactions as described in (A) but only showing PKC PKA and Akt. pAS; phosphorylated AS (n = 2). (C) AS was prephosphorylated with PKC (using cold ATP) then in vitro kinase reactions were carried out with GSK3 and CKII as in (A) (n = 1). A 0 10000 20000 30000 40000 50000 60000 70000 80000 PKCaPKAAMPKCKIIGSK3bPKGAkt KinaseActivity (CPM) Blank Kinase ** B Blank PKC Akt Blank PKA Akt pAS 117

PAGE 136

C 0 2000 4000 6000 8000 10000 12000 14000 Blank KinaseCorrected Activity (CPM) GSK3b CKII To test the dose-dependence of PKC and PKA phosphorylation of AS, in vitro kinase reactions were conducted with a constant level of PKC or PKA and varying AS concentrations (2.5 g to 12.5 g). Th e results demonstrate dose-dependent phosphorylation of AS by PKC that plateaus at 5 g of AS (Figure 15A). With PKA, the level of phosphorylation was linear up to 7.5 g AS (Figure 15B). In addition, there are numerous PKA and PKC motifs in the AS protein sequence, and this supports the in vitro data generated (Table 5). 118

PAGE 137

Figure 15: Dose Dependence of AS Phosphorylation. (A) In vitro kinase reactions carried out with a constant concentration of PKC and increasing concentrations of AS (2.5-12.5 g). Activity is represented as corrected CP M after subtracting blank (n = 1). (B) In vitro kinase reactions carried out with a constant concentration of PKA and increasing concentrations of AS (2.5-12.5 g). Activity is represente d as corrected CPM after subtracting blank (n = 3). 0 20000 40000 60000 80000 100000 120000 02.557.51012.5 AS Concentration (ug)Corrected Activity (CPM ) A 0 4000 8000 12000 16000 20000 24000 02.557.51012.5 AS Concentration (ug)Corrected Activity (CPM ) B 119

PAGE 138

Table 5: Possible Sites Phosphorylated by PKA or PKC Data generated with bioinformatics databases. Only sites that were positive hits in at least 2 of 4 databases are shown. Kinase Possible Sites PKA S2, S6, S115, T119, T174, S180, S219, S278, S352, S365, S396, T410 120 PKC S2, S6, T91, S115, T119, T174, T208, S219, T242, S352, S365, S396, T410 To further support the biological si gnificance of AS phosphorylation, we considered several candidate pathways th at could regulate AS phosphorylation. Insulin, VEGF and bradykinin pathways were all teste d. Preliminary data looked positive for all three pathways since all demonstrated th e ability to enhance AS phosphorylation. The data was strongest with vascular endothelia l growth factor (VEGF). Since both PKA and PKC have been linked to VEGF and NO pr oduction, this pathway was investigated further. In Chapter One, Figure 9A, we dem onstrated that VEGF significantly enhances NO production starting at 10 minutes and contin uing up to 2 hours. In order to assess whether PKA, PKC and eNOS were activated during acute stimulation of NO production by VEGF, we measured their phospho rylation after treatment with VEGF for 10 minutes. Indeed, PKA, PKC and eNOS had increased le vels of phosphorylation in response to VEGF, suggesting that these are key events that contributed to the enhanced production of NO (Figure 16A).

PAGE 139

121 Although eNOS is known to be activated during VEGF stimulation to enhance NO production, it is not known whether AS activity is also enhanced and necessary for increased NO production during VEGF stimula tion. To study this, the 2 hour time point was used in order to clearly distinguish VEGF-stimulated NO production under ASinhibited and uninhibited condi tions. BAECs were pre-treated for 1 hour with the AS inhibitor -methyl-DL-aspartic acid (MDLA, 10 mM ), an aspartic acid analogue that competitively inhibits AS [8]. The cells were then treated with VEGF (100 ng/ml) for 2 hours. As shown in Figure 16B, the VEGF-mediated increase in NO production was substantially diminished when endothelial cells were pre-tr eated with MDLA. When this experiment was repeated with a 10 minute tr eatment with VEGF, th e AS inhibitor MDLA essentially abolished both basal and VE GF-stimulated NO production (Figure 16C). To determine whether the requirement for AS activity for maximal VEGFmediated NO production was related to AS phos phorylation, BAECs were treated with 100 ng/ml VEGF for 10 minutes. Lysates were prepared and immunoprecipitations were carried out utilizing an AS antibody. S ubsequent western blotting with phosphoserine/threonine, phospho-PKC substrate and a phospho-PKA substrate antibody revealed that AS phosphorylation was enhanced by VEGF treatment (Figure 16D).

PAGE 140

Figure 16: VEGF is a Candidate Pathway for Regulating AS Phosphorylation. (A) Representative western blot (n = 3) of BAEC treated with VEGF (100 ng/ml, 10 minutes). Membranes were immunoblotted with phospho-PKC (pPKC ), total PKC phospho-PKA (pPKA), total PKA, phospho eNOS (peNOS), total eNOS and GAPDH (GAPDH). (B) NO assay of cells treated with MDLA alone (pre-t reated for 1 hour), VEGF alone (2 hours) or VEGF + MDLA. *** p < 0.00005; p < 0.05; n = 3 (C) Experiment conducted as in (B) with the exception that treatment with VEGF was for 10 minutes (n = 1). (D) Immunoprecipitation of cell lysates that were treated +/VEGF (100 ng/ml, 10 min) with an AS antibody (IP AS) or a same-species normal IgG (IP IgG) followed by probing with phospho-serine/threonine (pS/T), phospho-PKC substrate (pPKC Substrate) and phospho-PKA substrate (pPKA Substrate) antibodies. IgG = non-specific IgG band; pAS = phosphorylated AS band (n = 3). 122 A pPKA PKA Control VEGF peNOS GAPDH eNOS pPKC PKC B 0 500 1000 1500 2000 2500 3000 3500 ControlMDLAVEGFVEGF + MDLApmol nitrite/mg protein ***

PAGE 141

C -600 -400 -200 0 200 400 600 800 1000 1200 CMDLAVEGFVEGF + MDLApmol nitrite/mg protein D pS/T pPKA Substrate IP AS Control IP AS VEGF IP IgG IgG pAS pAS pPKC Substrate pAS 123

PAGE 142

124 Identification of Specific S ites of AS Phosphorylation: In order to assess the best methodology to pursue mass spectrometry id entification of AS phosphorylation, two approaches were taken to pur ify sufficient amounts of AS. For these experiments, we decided to utilize overexpressi on. Although this scenario is so mewhat artificial, the low stoichiometry of phosphorylation prompted us to take measures to maximize the possibility of detecting phosphorylated-A S peptides. The firs t purification method involved immunoprecipitating overe xpressed AS utilizing a V5 antibody to pull it down via its V5 tag. The expression levels of AS with his method were excellent and overexpressed AS was sufficiently purified (Fig ure 17A and B, respectively). In addition, AS, but not the empty vector control demons trated phosphorylation at serine residues both basally and in response to okadaic acid treatment (a phosphatase inhibitor). There was no apparent increase in phosphorylation with treatment (Figur e 17A). One possible problem with this methodology involves the f act that the heavy chain IgG (~55 kD) is inevitably pulled down (Figure 17B). Since it is close to the overexpressed AS molecular weight (~51 kD), IgG might interfer e with the identification of AS.

PAGE 143

Figure 17: Overexpression and Purificati on of AS Utilizing Immunoprecipitation (A) Immunoprecipitation was utilized to purify overex pressed AS-V5-His utilizi ng a V5 antibody. Blot compares samples that were treated with increasing concentrations of okadaic acid (OA) as indicated. Left side of blot shows immunoprecipitated samples (IP V5) and right side shows whole cell lysate (WCL) of samples expressing AS (AS-V5-His) or an empty vector. Top blot was probed with AS antibody and both overexpressed (AS: OE) and endogenous AS (AS: E) can be detected. Bottom blot was probed with a phospho-serine (pS) antibody. (B) Image of duplicate gel that was coomassie stained. Red box indicates AS (51 kD). Black box indicates non-specific heavy chain IgG band (~50-55 kD). Protein marker is indicated by arrow. Post-IP Sup IgG Control Control OA 20 nM Oa 100 nM Control OA 20 nM Oa 100 nM Control OA 20 nM Oa 100 nM Control OA 20 nM Oa 100 nM AS-V5-His Empty Vector AS-V5-His Empty Vector IP V5 WCL AS OE AS E pS A 125 B Protein marker

PAGE 144

126 This prompted us to test a second method that involved utilizing Ni-NTA magnetic beads to affinity purify overexpres sed AS via its 6X-His tag. This method allowed for excellent enrichment and purific ation of the overexpressed AS (Figure 18A and B). It also allowed for a significant amount of endogenous AS to co-purify (Figure 18A). Considering the effectiv eness of this method and the lack of IgG interference, it was chosen for downstream mass spectrometry experiments. In addition, as shown in Chapter One, Figure 4, AS overexpression le ads to enhanced NO production, suggesting that overexpressed AS functions similarly to the endogenous enzyme. Overall, this data supports the use of the overexpression syst em to identify specific sites of AS phosphorylation.

PAGE 145

Figure 18: Overexpression and Purifi cation of AS Utilizing Ni-NTA. (A) Ni-NTA agarose was utilized to purify overexpressed AS-V5-His. Blot compares samples th at were treated with increasing concentrations of okadaic acid (OA) as indicated. Left side of blot shows purified samples (NI-NTA Purified) and right side shows whole cell lysate (WCL) of samples expressing AS (AS-V5-His) or an empty vector. Top blot was probed with AS antibody and both overexpressed (AS: OE) and endogenous AS (AS: E) can be detected. Bottom blot was probed with a phospho-serine (pS) antibody. (B) Image of duplicate gel that wa s coomassie stained. Red box indicates AS (51 kD). Protein markers are indicated by arrows. AS-V5-His Empty Vector AS-V5-His Empty Vector Ni-NTA Purified WCL Control OA 20 nM Oa 100 nM Control OA 20 nM Oa 100 nM Control OA 20 nM Oa 100 nM Control OA 20 nM Oa 100 nM AS OE AS E pS A 127 Protein marker Protein marker B

PAGE 146

128 In order to identify spec ific sites of AS phosphorylation, AS was overexpressed and cells were treated with okadaic acid, insulin, bradykinin, and insulin + bradykinin. AS was purified utilizing Ni-NTA agarose magnetic beads and cellular protein was separated utilizing SDS-PAGE. The band corresponding to overexpressed AS was excised and prepared for liquid chromatogr aphy-tandem mass spectrometry (LC-MS/MS) analysis. We have thus far identified the following sites of AS phosphorylation: T131, Y133, S134, S180, S189, Y207 and S328. The sites that have been identified on several occasions are T131, S180 and S189 while th e remaining sites have not yet been confirmed. Assessment of the cumulative data did not provide clear determination as to whether any particular treatment enhanced phosphorylation at a specific site. We did note that the patterns of AS phosphorylation are indeed dynamic. Attempts were made to relatively quantitate the phos phorylation of AS +/differe nt treatments utilizing data from the liquid chromatography phase to enhance peptide identification via mass spectrometry. This data was also inconclusive However, there were trends noted that suggested changes in phosphorylation pattern s. In addition, the peptide 131-TCYS-134 contains multiple possible modifications. Their close proximity makes it impossible to determine which site is actually modified (i.e. T131 vs. S134). Figure 19 is an example of the data generated via LC-MS/MS. The spect rum at the top shows that under control conditions, T131 is not phosphor ylated. The spectrum below it indicated that upon okadaic acid treatment, T131 becomes phosphor ylated. The mass to charge ratio (m/z) difference of a peptide phosphorylated by a se rine or threonine modification is ~80. The m/z difference of 11 between the non-phos phorylated vs. phos phorylated T131 was derived by the fact that the non-phosphorylated peptide was alkylated (m/z = 57). The

PAGE 147

m/z difference between the non-phosphorylat ed and phosphorylated peptide with a charge of +2 was calculated as follows: (80-57)/2 = 11. Figure 19: Identification of AS Phosphoryl ation Sites Utilizing Liquid ChromatographyTandem Mass Spectrometry. AS was overexpressed and cells were treated with okadaic acid (100 nM) for 30 min. After purification and SDS-PAGE, the band corresponding to overexpressed AS was excised digested, and subjected to liquid chromatography-tandem ma ss spectrometry. Figure shows mass spectra of control (top) and okadaic acid (bottom) treated samples. In the control samples, T131 is not phosphorylated while in the okadaic acid samples, T131 is phosphorylated. 129 Control Okadaic Acid

PAGE 148

130 Although AS is the focus of this wo rk, we did identify some novel eNOS phosphorylation sites that are worth noting. These experiments were carried out by immunoprecipitating eNOS as described in Ch apter Three, Page 166, and subjecting the eNOS band to LC-MS/MS to identify phosphor ylation sites. The novel sites of bovine eNOS phosphorylation were: T60, T62, T389, and S485. Since the focus of this work is se rine/threonine phosphorylation, further investigation into biologi cal significance was conducted for only those modifications. First, to determine the evolutionary relevance of the sites, a multiple sequence alignment was conducted. As seen in Figure 20, T131, S134, S180, S189 and S328 are 100% conserved among mammals with the exception that humans have a serine at position 131 while other mammals have a threonine. The si tes that were completely conserved even down to insects and E. coli were S180 and S189. The only minor substitution was a threonine at position 189 in E. coli. Overall, the mammalian AS sequences depicted in the figure are about 95% identical. Sequence iden tity then declines with the lowest being 25% for the E. coli sequence compared to human or bovine (per ClustalW analysis). With such a low level of conservation among some of the non-mammalian species, the persistence of phosphorylatable residues at some of the positions strongly suggests biological importance.

PAGE 149

Figure 20: Multiple sequence alignment of AS Phosphorylation Sites. Panels show the peptide containing T131 and S134 (A), S180 and S189 (B) or S328 (C). Sites are highlighted in blue. Human FELSCYSLAPQIKVIA 143 131 Bovine FELTCYSLAPQIKVIA 143 Rat FELTCYSLAPQIKVIA 143 A Mouse FELTCYSLAPQIKVIA 143 Dog FELTCYSLAPQIKVIA 143 Frog FELTCYSLYPEVKIIA 144 Zebrafish FELTCYALYPQVQVIA 142 Fruit Fly FELCAYALKPDLKIIA 142 Yeast FELSFYALKPDVKCIT 142 E. coli FYRYGLLTNAELQIYK 155 Human KNPWSMDENLMHISYE 191 Bovine KNPWSMDENLMHISYE 191 Rat KSPWSMDENLMHISYE 191 B Mouse KSPWSMDENLMHISYE 191 Dog KNPWSMDENLMHISYE 191 Frog KDPWSMDENIMHISYE 192 Zebrafish KAPWSMDANLMHISYE 190 Fruit Fly ATPWSTDANILHISYE 190 Yeast AKPWSTDENQAHISYE 190 E. coli EKAYSTDSNMLGATHE 203 Human LKFAELVYTGFWHSPE 330 Bovine LKFAELVYTGFWHSPE 330 Rat LKFAELVYTGFWHSPE 330 C Mouse LKFAELVYTGFWHSPE 330 Dog LKFAELVYTGFWHSPE 330 Frog QRFAEQIYNGFWYSPE 331 Zebrafish IKFSELIYNGFWFSPE 329 Fruit Fly DRMADYVYNGFWFSPE 330 Yeast PNYSRLIYNGFLLHPE 334 E. coli RQLGRLLYQGRWFDSQ 340

PAGE 150

To further elucidate the biological role of these sites, an in silico experiment was conducted to identify putative kinases. As seen in Table 6, several important kinases were identified as possible kinases for the id entified sites including PKA, AMPK, CKII and GSK3 Interestingly, PKC motifs were not found at any of the AS phosphorylation sites identified so far. Table 6: Putative Kinases fo r Identified AS Serine/Threo nine Phosphorylation Sites. Data was generated based on motif searches with Prosite or via bioinformatic databases. RESIDUE PUTATIVE KINASES T131 AMPK CKI & CKII MAPK S134 CKI S180 PKA CKII PKG S189 CKII MAPK S328 GSK3 CKII MAPK Another methodology to explore biological significance involves utilizing the 3dimensional structure of an enzyme to asse ss possible structure-func tion relationships. As indicated in Table 7, each of the identified AS serine/threo nine phosphorylation sites is linked to physiologically relevant path ways. T131 and S134 are located in the Nterminal, nucleotide binding domain of AS, which has been shown to be involved in conformational change during catalysis in the E. coli enzyme. T131/S134 are also 132

PAGE 151

adjacent to cysteine 132, a site that is revers ibly nitrosylated in vascular smooth muscle and functions as a negative regulator of AS enzyme activity under conditions of excessive NO production. S180 is in the N-termin al domain but is part of the catalytic cleft. It is hypothesized to interact with ATP {per e-mail conversation with Dr. Jonas Uppenberg [9]}. Interestingly, S180 is mutated in citrullinemia (S180 N), a disease characterized by diminished AS function. S189 is also in the N-termin al domain as part of the catalytic cleft and hydrogen bonds with citrulline. S328 is loca ted in the catalytic domain. Additionally, it is adjacent to the AS caveolin-1 binding motif (discussed in Chapter Three), suggesting a role in regulati ng the interaction between AS and caveolin. Table 7: Hypothesized Biological Significance of AS Phosphorylation. Identified AS serine/threonine phosphorylation sites, their hypothesized relevance based on the literature and 3D structure analysis and the nu mber of times that each site was identified in bioinformatic databases. RESIDUE HYPOTHESIZED BIOLOGICAL ROLE DATABASE FREQUENCY T131/S134 Located in the nucleotide domain which is hypothesized to be involved in conformational change upon catalysis. Near cysteine 132, a site known to be nitrosylated under conditions of excess NO production. This site inhibits AS activity. It is possible that the phosphorylation and nitrosylati on serve as on-off switches in response to cellular NO needs. 3/1 S180 Located in the catalytic cleft and putatively binds ATP This site is known to be mutated in citrullinemia suggesting that it is important for AS activity. 4 S189 Also located in the catalytic cleft. Forms a hydrogen bond with citrulline so it is important for AS activity. 2 S328 This site is located in the catalytic domain It is in close proximity to T131. Additionally, it is adjacent to the AS caveolin binding motif suggesting a role in regulating this putative interaction. 2 133

PAGE 152

Figure 21 shows the 3-dimensional struct ure of an AS monomer, dimer and tetramer and illustrates its intricate complexity. The location of the identified phosphorylation sites with respect to the 3-dime nsional structure of AS is depicted in Figure 22. Figure 21: The AS 3-Di mensional Structure. AS monomer (A), homo-dimer (B) and homo-tetramer (C) as illustrated by the Structural Genomics Consortium iSEE software (A) or generated by the USF Department of Chemistry (B, C). A B C 134

PAGE 153

Figure 22: Structure-Function Relationships of AS Phosphorylation Sites. (A) AS 3D structure (monomer) with location of identified phosphorylation sites in relation to domains (modified from Structural Genomics Consortium iSee software). The N-terminal domain is in green, the catalytic domain is in orange, the C-terminal domain is in gray and the protruding loop that links the Nterminal and catalytic domains is in purple. (B) AS 3D structure (monomer) with location of identified phosphorylation sites in relation to subs trates (citrulline and aspartate) (modified from ViewerLite-generated structure) Arrows point to the residues and substrates. Phosphorylated residues are identified in ball and stick form in bl ack. Aspartate and citrulline are in CPK form (gray and purple, respectively). A 135

PAGE 154

T134 S180 S189 Citrulline Aspartate S131 S328 B Mechanism of AS Regula tion by Phosphorylation: To understand the role of specific AS phosphorylation sites on enzyme activity, phospho-null (S/T A) and phospho-mimetic (S/T D) variants were generated for T131, S180 and S189, the 3 sites that were confirmed several times by mass spectrometry. First, th e effect of alanine mutations was tested. As shown in Figure 23A, preventing phosphor ylation at any of these sites significantly reduced the abili ty of endothelial cells to produce NO. Mimicking constitutive phosphorylation at t hose sites yielded some recovery of NO production, especially with the S180D variant, but NO levels did not reach those of the wild-type enzyme (Figure 23B). A repeat experiment where the phospho-null and phospho-mimetic variants were compared head-to-head confirmed the observations made when the variants were test ed individually (Figure 23C). Since the S180 and S189 have 136

PAGE 155

such essential catalytic functions, a double variant was created where both sites were mutated to alanine. As shown in Figure 23D the double mutation did not further diminish NO production as compared to the single mutation. Figure 23: Role of T131, S180 and S189 on Endothelial Nitric Oxide Production. BAECs were transiently transf ected with wild type (WT) empt y vector (EV) or variants (T131A, T131D, S180A, S180D, S189A, S189D, S180A189A). (A) Comparison of WT AS to phospho-null alanine variants (n = 2). (B) Comparison of control (Cuntransfected), EV and WT AS to phospho-mimetic (aspartic aci d) variants (n = 1). (C) Head to head comparison of EV and WT versus phospho-null and phospho mimetic variants (n = 2). (D) Comparison of control, empty vector and wild type AS to single and double alanine mutants at p ositions 180 and 189 ( n = 3 ) p < 0.02; ** p < 0.007 A 0 0.5 1 1.5 2 2.5 WTT131AS180AS189A ConstructFold NO produce d Vehicle Expressed 137

PAGE 156

B 0 1000 2000 3000 4000 5000 6000 CEVWTT131DS180DS189D Constructpmol nitrite/mg protein C 0 0.5 1 1.5 2 2.5 3 EVASS180AS180DS189AS189D ConstructFold NO Produce d 138

PAGE 157

0 1000 2000 3000 4000 5000 6000 7000 8000 9000 CEVWTS180AS189AS180A189A Constructpmol nitrite/mg protein ** ***** D The work with the AS phospho-null a nd phospho-mimetic variants led to an important observation. Consistently, when BA ECs were transiently transfected with wild-type AS, there was a high de gree of cell loss that incr eased over time. Theoretically, this was due to the role of AS in maintain ing endothelial cell viability by maintaining appropriate amounts of NO as demonstrated in Chapter One. The phospho-null variants of AS never demonstrated any significant cell lo ss or even visible cell stress. This highly suggested that all 3 sites ha ve a role in AS function. Another approach that was utilized to assess biological significance of AS serine/threonine phosphorylati on involved the use of comput ational modeling. Since the human crystal structure of AS was solved re cently, we now have an invaluable tool to study structure-function relationships. The theo ry behind this line of experiments was that certain amino acids may enhance or interfer e with substrate binding while others might be more accessible to modification by phosphor ylation. To test thes e theories, all the 139

PAGE 158

140 identified serine or threonine phosphoryl ation sites [T131 (S131 in human structure), S134, S180, S189 and S328] were modeled plus or minus the phosphate modification and plus or minus substrates. The Molecular Dyna mics method was then utilized to measure substrate binding affinity. Serines 180 and 189 were found to flank citrulline. Phosphorylation at either of these sites led to significantly diminished affinity for substrates. In addition, in the tetrameric stru cture those sites are co mpletely buried in the active site (Figure 24). S131, S 134 and S328 lie in close proximity in the 3-dimensional structure (Figure 25). In the dimer, these 3 serines are partially covered by the "free" Cterminal helix that is part of the dimer-dimer interface. Although th is region is partially buried, there is surface exposure. S328 has its oxygen facing away from solvent. The oxygen of S131 is hydrogen bonded to H327. S134 has its proton hydrogen-bonded to the backbone carbonyl of H327 leaving its oxygen exposed to solvent and ready for nucleophilic attack. In Figure 26, a close up view of these sites is shown that demonstrates that S134 and S328 are more accessible than S131. S131 has its hydroxyl group buried by the loop where S328 is located. In this figure it is also evident that S131 and S134 are in the outer portion of an alpha helix. In summary, the in silico modeling of AS plus or minus phosphorylation at identi fied sites revealed that the most likely candidates for modification by phosphor ylation are S131, S134 and S328.

PAGE 159

Figure 24: Close Up of Human AS Active Site. Serines 180 (left) and 189 (right) are shown in yellow Aspartate (left) and citrulline (right ) are shown in green. ATP would bind in the empty space in the lower left quadrant of the figure. 141

PAGE 160

Figure 25: In Silico Modeling of AS Residues with Good Accessibility for Modification by Phosphorylation. Close up of human AS structure showing S131 (bovine T131), S134 and S328. On top of the figure is the free c-terminal helix that forms the dimer-dimer interface. This helix partially buries these sites. 142

PAGE 161

Figure 26: Close-Up View of S131, S134 and S328. Carbon residues are labeled in gold. S134 and 328 are on the upper left and upper right. S131 is on the lower left. 143

PAGE 162

144 Discussion In this Chapter we identified the firs t biologically significan t post-translational modification of AS in the endothelium. The majo rity of AS regulation has been studied at the level of transcription/translation [10] and very little is known about the posttranslational regulation of AS. The work described in this Chapter is the first comprehensive investigation of one post-tr anslational mechanism for regulating AS function. Although there have not been any post-translational modifications of AS identified in endothelial cells, the post-tra nslational modification of AS was recently noted by Hao et. al [11]. They found that AS is nitrosylated and inact ivated in vascular smooth muscle under conditions of excess NO production via iNOS, suggesting that the activity of AS is at l east partially responsible for sensi ng cellular NO levels and adjusting NO output accordingly. This addition of an NOderived nitrosyl group to the AS protein sequence also suggests a feedback mechanis m where NO levels that exceed a certain threshold lead to a decrease in NO output In addition, in a ph ospho-proteomics study utilized to identify phosphoproteins in HeLa cells, AS was found to be phosphorylated (S352) [12]. However, the scope of the paper was such that the biological relevance of AS phosphorylation was not explored. In our work, the finding that AS is an endogenous endothelial phosphoprotein was enhanced by the identification of PKA and PKC as AS kinases and the regulation of this phosphorylation by VEGF.

PAGE 163

145 The importance of tight control of NO production is highlighted by vascular endothelial cells where virtua lly all phenotypic properties ar e related to NO bioactivity. Because of this, cardiovascular risk factors often mediate their deleterious effects by compromising these controls which leads to endothelial dysfunction. The finding that PKA and PKC phosphorylate AS in addition to their known phosphorylation and activation of eNOS [13-18] suggests that during VEGF stimulation, these kinases act coordinately on eNOS and AS via phos phorylation to enhance NO production in endothelial cells. The specificity of the phosphorylation of AS by PKA and PKC is supported by the fact that Akt, an essentia l kinase in regulating NO production [19-22], did not phosphorylate AS. Additionally, the substrate-dependence of the reaction and the multiple motifs found in the AS sequence suggest that PKA and PKC are bona fide AS kinases. The specific effects of PKA and PKC on AS function are not clear. Although the evidence at this time compels us to believe th at both kinases activate AS, this may in fact not be the case. First, we have to consider the limitations of the tools we utilized to assess whether PKA or PKC -mediated AS phosphorylation was indeed enhanced by VEGF. Second, we need to be able to test the effect of each kinase on the sp ecific activity of AS. Finally, we need to know specifically which sites are phosphorylated by these kinases. All this work is currently underway. Although there is still a subs tantial amount of work that needs to be done to acquire a full understanding of PKA and PKC -mediated AS phosphorylation, the

PAGE 164

146 literature does support our data and hypothese s. So far, the story with PKA and NO production seems fairly clear. Under many c onditions (VEGF, bradykinin, shear stress), PKA increases eNOS phosphorylation at ac tivating sites [13-16]. Our level of understanding of the role of PKC (and PKC in general) in regulating NO production is not so clear. In many instances, PKC, and PKC specifically, decrease NO production by a variety of mechanisms [14, 17]. One notable mechanism involves the increase of eNOS phosphorylation during VEGF stimulation by PKC at an inactivating site thus reducing its activity [14]. In fact, in that particular study, PKA and PK C acted reciprocally. On the other hand, a couple of examples link PKC to enhanced NO production. First, in vivo studies demonstrated th at overexpression of PKC in rat femoral arteries results in an increase in eNOS-mediated blood flow [18]. In another instance, PKC was shown to be important for the maintenance of vascular integrity during chronic inflammation. This involved the activation of VEGF by decay-a ccelerating factor [23]. In addition, PKC enhances eNOS expression [24]. Therefore, it is evident that the phosphorylation of AS by PKC and PKA is a very significant finding that requires careful and extensive investigation. There were several other kinases that were not able to phosphorylate AS in vitro in our initial studies, but wh ich warrant continued investigation. AMPK did demonstrate some increase in 33P ATP incorporation, but it was not found to be significant. AMPK has an extremely important role in regul ating NO production and energy metabolism [25, 26], two functions it shares with AS. For example, AMPK activates eNOS and thus increases NO production in response to ad iponectin, shear stre ss insulin and VEGF

PAGE 165

147 stimulation [26-29]. Also, AMPK helps to normalize some of the endothelial defects caused by high glucose by activating eNOS through interactions with HSP90 [30]. AMPK also protects endothelial cells agains t high-glucose induced apoptosis [31]. Since AS is involved in anti-apoptotic cascades in the endothelium, part of the protection by AMPK could be mediated by AS activation. In addition, the pathways we have linked to ASinsulin, VEGF, TNF ceramide and PPAR have also been linked to AMPK function [28, 32-35]. We did try two different combinations of AMPK catalytic and regulatory subunits for the in vitro kinase screens and neither gave a significant result. There is the possibility that some thing present or absent in the in vitro reaction is preventing significant phosphor ylation of AS by AMPK. We do hypothesize that if AMPK phosphorylates AS, it is like ly an activating modification. Also, CKII and GSK3 have important roles in regulating NO production but failed to demonstrate the ab ility to phosphorylate AS in vitro For example, GSK3 is a downstream target of Akt, a major regulat or of eNOS function [36, 37]. In addition, GSK3 regulates angiogenesis in endotheli al cells [38, 39]. In one study, it was determined that GSK3 is downstream not just of Akt signaling, but also PKA and MAPK [39]. The mechanism by which GSK3 regulates angiogenesis involves a downregulation of matrix attachment and mi gration [39]. Consider ing the regulation of VEGF-mediated NO production by AS activit y, it is certainly possible that GSK3 phosphorylates AS. Thus far, eNOS has not been found to be phosphorylated by GSK3

PAGE 166

148 In addition, CKII has been shown to phosphorylate calmodulin and this inactivates eNOS [ 40]. Furthermore, CKII phosphorylat es and inactivates protein phosphatase 2A and leads to decreased SP-1 binding to the eNOS promoter and ultimately to decreased eNOS expression [41]. CKII has not been shown to phosphorylate eNOS. An interesting link betw een CKII and NO metabolism that has not been explored is the fact that it phosphorylates PKA and HSP90 [42, 43], two important proteins that regulate NO production. Often times, both GSK3 and CKII require their s ubstrates to first be phosphorylated by another kinase before they can phosphorylated th eir target [44-46]. We believe that these two kinases may ind eed phosphorylate AS, but only when priming phosphorylation occurs first. We did a ttempt to pre-phosphorylate AS with PKC and then tried the in vitro kinase reactions with CKII and GSK3 We also obtained negative results, suggesting that anot her kinase needs to pre-phosphor ylate AS at sites that are specifically adjacent to the mo tifs recognized by CKII and GSK3 Considering the negative regulation of eNOS by both GSK and CK II, it is possible that these kinases may act to diminish AS activity. Finally, although AS was also not a good substrate for PKG in vitro, we did identify this kinase as a possible interac ting partner with AS and eNOS in BAECs (discussed in Chapter Three). PKG is associated with multiple functions of the endothelium including vasodi lation, angiogenesis, improvement of vasodilation during recovery from heart failure and the up-regul ation of the mitochondr ial oxidative stress

PAGE 167

149 protection system [47-53]. In addition, P KG co-localizes with eNOS in endothelial caveolae [54], suggesting that it is part of the functional signaling microdomain necessary for citrulline-NO cycle function. Since we ha ve identified AS to be part of this microdomain ([55]; See also Chapter Three), it is quite possible that PKG has direct effects on endothelial AS function. Although most of what is known about PKG regulation of vascular health occurs specifically in smooth muscle, PKG does phosphorylate eNOS at S633 and S1179 [56]. Thus there is still a possibility that our in vitro kinase screening results are simply false negatives and that there are in fact mechanisms of PKG regulation of AS that occur specifically in the endothelium. The biological importance of AS phosphoryl ation by identification of AS kinases is strengthened by our identif ication of VEGF as one pa thway for post-translational regulation of AS. Our finding that AS activity is necessary for maximal activation of NO production by VEGF draws a clear link be tween AS function and the endotheliumspecific biological roles of VE GF such as angiogenesis and vasodilation [28, 52, 57-60]. The decrease seen with the data generated at the two hour time point represented a state where continued stimulation by VEGF allowe d for sufficient activation of eNOS and possibly the utilization of co mpensatory mechanisms so that NO could still be produced at levels significantly above controls. In comparison, when this experiment was conducted at the 10 minute time point, which re presented and acute s cenario of AS and eNOS activation, MDLA alone led to a dramatic reduction in basal NO production and abolished the VEGF-mediated increase in NO production. This repr esents a situation where there is insufficient time for additional compensatory mechanisms to emerge in an

PAGE 168

150 effort to restore the cells capability to respond to the VEGF stimulation. This also highlights just how critical acu te activation of AS is for a suitable response to simulation. Furthermore, the implication that VE GF stimulates AS phosphorylation is significant due to the known mechanisms by which VEGF leads to activation of eNOS with concomitant increases in NO production [14, 52, 61, 62]. The data showing an increase in AS phosphorylation signal w ith a phospho-PKA and phospho-PKC substrate antibodies suggests that VE GF leads to the phosphoryla tion of AS by PKA and PKC Although not shown due to difficulty in generatin g a clear image, it do es seem that there are several bands that are near the AS mo lecular weight that are phosphorylated by PKA and PKC in response to VGEF and other treatm ents. Additionally, when AS is purified via immunoprecipitation, two bands near the AS molecular weight are noted when a gel is coomassie stained. Therefore, it is possi ble that there are multiple phosphorylated AS species. Considering that VEGF and eNOS expr ession and function are diminished in diabetes and cardiovascular disorders [63, 64] the inclusion of AS in this regulatory scheme suggests that there are multiple unexpl ored roles of AS in these diseases. This work is supported by the findings of Shen et al. demonstrating that AS activity is essential for endothelial NO pr oduction mediated by eNOS (not iNOS) and that inhibiting its activity with MDLA diminishes maxi mal NO production by the calcium ionophore A23187 [65]. This supports both the functi onal importance of AS for endothelial NO production and links AS to calcium signaling, which is part of the mechanism by which

PAGE 169

151 VEGF mediates NO production [ 61]. Additionally, there is re ciprocal regulation between NO and VEGF designed to restore homeostasis [66] leading to th e distinct possibility that there are mechanisms that will diminish th e response of AS to VEGF. Although eNOS is critically important for the regulation of vascular NO produc tion, our work demonstrates that the function of the citrulline-NO cycle as a whole is essential and warrants further study to provide a more global understa nding of such an important system. The regulation of NO production via revers ible phosphorylation is exemplified extensively by eNOS. There are at least 5 known sites of eNOS serine/threonine phosphorylation that eith er activate or inactivate its func tion. These sites are responsive to a variety of stimuli [67-69]. We have noted a similar mechanism for AS. Our molecular biology studies demonstrated time and time again, that AS is phosphorylated basally and that this changes, qualitatively or quantita tively, upon stimulation. Our mass spectrometry analysis identified 5 sites of serine/threonine phos phorylation and 2 of tyrosine phosphorylation. The phosphorylation at these sites was dyna mic and responsive to stimuli. One example was the enhancemen t of AS phosphorylation at T131 in response to okadaic acid, a serine/threonine phosphatase inhibitor. Specifical ly, okadaic acid has been shown to diminish the activation of ceramide activated protein phosphatases [70], which suggests a link between ceramide and AS activity in addition to its effects on AS expression demonstrated in Chapter One, Figure 10. If indeed there is a direct relationship between ceramide and the regula tion of AS on multiple levels, this would mimic very closely what is known about ceramide and eNOS [71, 72].

PAGE 170

152 There were significant difficulties in obt aining reproducible and reliable data related to quantitation of AS phosphoryla tion due to the low stoichiometry of phosphorylation, the labile nature of these modifications and possibly due to the methods used to overexpress and purify AS. At times the precise site that was actually phosphorylated was ambiguous due to close prox imity of more than one phosphorylated site. These types of problems are common a nd this is precisely why these types of projects are known to be extremely difficult to accomplish [73]. We do know, without a doubt, that AS is indeed phosphorylated and we will continue to optimize our methods so that characterization of th e function of each relevant phosphorylation site can be achieved. It is possible that one way to solve these issues is to assess organelle-specific AS phosphorylation by, for example, enriching caveolar fracti ons. There may be distinct areas of the cell where phosphorylation is mo re prominent and assessing this in whole cell preparations may be diluting the results. This will be further discussed in Chapter Three. Despite the limitations our collective find ings strongly suggest that AS is regulated by a complex interplay of phos phorylation-dephosphorylation evens that generate a cellular barcode that defines AS function. Based on our in silico studies, it is also likely that there are several kinases that phosphorylate AS, mediating the di fferential signaling necessary to maintain tight control over endothelial NO production. Table 6 lists possible kinases that can phosphorylate AS at indentified phosphorylation si tes. Several of them have been discussed in previous sections: PKA (S180), AMPK (T131), GSK3 (S328), and CKII (T131, S180, S189, S328). This section will focus on some of the other kinases identified.

PAGE 171

153 Besides CKII, CKI was also found to be a putative kinase for T131 and S134. There have not been any direct studies of CKI in the nitr ic oxide system. CKI and CKII have both overlapping and distinct substrate specificities, although they each recognize a different motif in their targets [74]. CK I does phosphorylate glycogen synthase which implies a role in glucose metabolism [75]. Thus, a link to nitric oxide function is possible. Mitogen activated protein kinase (MAPK) family members were identified as additional putative kinases for T131, S189 and S328. Members of the MAPK/extracellular signal-regulated kinase (ERK) family are known to have multiple roles in regulating NO production. One exampl e is an increase in eNOS expression by angiotensin II which is mediated by MAPK [76]. In addition, MAPK activates PPAR in response to NO and may represent a cardiop rotective mechanism [77]. In addition, MAPK activates eNOS by increasing phosphorylation at S1177 and decreasing phosphorylation at T495 in response to black tea polyphenols [78]. ERK is also involved in mediating the pro-angiogenic effects of NO [79]. Thus, it is possible that at least one MAPK family member is involved in regul ating AS activity by phos phorylation. Overall, the combination of in vitro and in silico experiments to identify AS kinases strongly suggested that multiple kinases control the dynamic phosphorylation of AS. Our investigation of the mechanism of AS phosphorylation by site-directed mutagenesis suggested that T131, S180 and S 189 are all important for AS function. This data combined with the in silico modeling of the human AS crystal structure with or

PAGE 172

154 without phosphorylation at all th e serine/threonine sites iden tified by mass spectrometry led to additional implications. First, T131, S134 and S328 are the s ites with the highest possibility of being phosphorylated. This certa inly supports the mutagenesis data with T131 that showed significantly diminish ed NO production in the phospho-null variant and partial recovery with the phospho-mimetic va riant. It will be interesting to correlate these findings, future molecular biology e xperiments and the information we have on bioinformatic database frequency. This w ill allow us to gauge how bioinformatic database frequency relates to true biological significance. In addition, the decrease in AS-med iated NO production seen with S180 and S189 is most likely due to the di rect interactions of these si tes with substrates [9]. If phosphorylation can indeed occur at these sites, then the effects we saw are likely due to alterations in substrate binding. Alternatively, even if phosphorylation does not actually occur at those sites, the data we obtained is li kely due to the fact that the alanine mutants are missing the hydroxyl side chain of the seri ne and again this w ould alter substrate binding and inhibit NO producti on. Further, although the qua ternary structure of AS suggests that these sites are not accessible for phosphorylation, it is possible that they are phosphorylated only in the monomeric conf iguration. Perhaps phosphorylation of AS monomers is a mechanism that triggers the formation of the active tetramer and this has mechanism has not yet been de fined. In addition, even seem ingly inaccessible sites may become accessible upon conformational change due to physical interactions with kinases or proteins that can deliver kinases to substrates such as heat shock proteins.

PAGE 173

155 Finally, an intriguing possibi lity related to AS phosphoryl ation is the proximity of T131 and S134 to the identified nitrosylaiton of AS at C132 in vascular smooth muscle [11]. The nitrosylation of this site in endothelium and the putative cross-talk between AS phosphorylation and nitrosylation have not been studied. Furthermore, in endothelial cells, eNOS has been found to be basally nitr osylated (inactive) then de-nitrosylated (active) in response to VEGF s timulation [80]. It is a possibility that a similar mechanism regulates AS due to the important link we made between VEGF signaling and AS function. There are two other types of post-transl ational modifications that were not addressed experimentally in this work, but are worth mentioning. First, as discussed further in Chapter Three, acylation is an important mechanism for the subcellular targeting of eNOS with important functional consequences [81]. An in silico search for acylation motifs is the AS sequence suggested that these modifications do not occur in AS. Further, it has recently been determined that eNOS that is phosphorylated at serine 1177 and thus active, is inactivat ed by glycosylation at this s ite in diabetes [82]. A global investigation of proteins modified by O-glycosylation f ound that AS can be glycosylated [83]. Thus, it is clear that the barcode of dynamic post-transl ational modifications of AS including phosphorylation, nitr osylation and glycosylation illustrate an intricate mechanism of AS regulation that defines its tissue-specific functions and highlights the biological relevance of this enzyme.

PAGE 174

156 In summary, this Chapter has presented th e first comprehensive exploration of AS regulation by dynamic serine/threonine phosphorylation. Although deciphering and understanding the phosphoproteome is one of th e most complex tasks in proteomics [73], the findings we have to this point have greatly increas ed our understanding of AS regulation and will open up multiple avenues of important experimentation with far reaching implications for vascular biology. References [1] T. Michel, G. K. Li, and L. Busconi, P hosphorylation and subcellu lar translocation of endothelial nitric oxide syntha se, Proc Natl Acad Sci U S A 90 (1993) 6252-6256. [2] L. C. Pendleton, B. L. Goodwin, L. P. So lomonson, and D. C. Eichler, Regulation of endothelial argininosuccinate synthase expression a nd NO production by an upstream open reading frame, J Biol Chem 280 (2005) 24252-24260. [3] N. Blom, S. Gammeltoft, and S. Brunak, Sequence and structure-based prediction of eukaryotic protein phosphoryl ation sites, J Mol Biol 294 (1999) 1351-1362. [4] R. Linding, L. J. Jensen, G. J. Ostheimer, M. A. van Vugt, C. Jorgensen, I. M. Miron, F. Diella, K. Colwill, L. Taylor, K. Elder, P. Metalnikov, V. Nguyen, A. Pasculescu, J. Jin, J. G. Park, L. D. Samson, J. R. Woodgett, R. B. Russell, P. Bork, M. B. Yaffe, and T. Pawson, Systematic discovery of in vivo phosphorylation networks, Cell 129 (2007) 1415-1426. [5] R. Amanchy, B. Periaswamy, S. Mathiv anan, R. Reddy, S. G. Tattikota, and A. Pandey, A curated compendium of phos phorylation motifs, Nat Biotechnol 25 (2007) 285-286. [6] Y. Xue, F. Zhou, M. Zhu, K. Ahmed, G. Chen, and X. Yao, GPS: a comprehensive www server for phosphorylation sites prediction, Nucleic Acids Res 33 (2005) W184187. [7] B. Bodenmiller, L. N. Mueller, M. Mueller, B. Domon, and R. Aebersold, Reproducible isolation of distinct, overla pping segments of the phosphoproteome, Nat Methods 4 (2007) 231-237.

PAGE 175

157 [8] S. Ratner, Enzymes of arginine and ur ea synthesis, Adv Enzymol Relat Areas Mol Biol 39 (1973) 1-90. [9] T. Karlberg, R. Collins, S. van den Be rg, A. Flores, M. Hammarstrom, M. Hogbom, L. Holmberg Schiavone, and J. Uppenberg, Structure of human argininosuccinate synthetase, Acta Crystall ogr D Biol Cr ystallogr 64 (2008) 279-286. [10] A. Husson, C. Brasse-Lagnel, A. Fairand, S. Renouf, and A. Lavoinne, Argininosuccinate synthetase from the ur ea cycle to the citrulline-NO cycle, Eur J Biochem 270 (2003) 1887-1899. [11] G. Hao, L. Xie, and S. S. Gross, Argininosuccinate synthe tase is reversibly inactivated by S-nitrosylation in vi tro and in vivo, J Biol Chem 279 (2004) 36192-36200. [12] K. Imami, N. Sugiyama, Y. Kyono, M. Tomita, and Y. Ishihama, Automated phosphoproteome analysis for cultured can cer cells by two-dimensional nanoLC-MS using a calcined titania/C18 biphasic column, Anal Sci 24 (2008) 161-166. [13] Y. C. Boo, J. Hwang, M. Sykes, B. J. Michell, B. E. Kemp, H. Lum, and H. Jo, Shear stress stimulates phosphorylation of eNOS at Ser(635) by a protein kinase Adependent mechanism, Am J P hysiol Heart Circ Physiol 283 (2002) H1819-1828. [14] B. J. Michell, Z. Chen, T. Tiganis, D. Stapleton, F. Katsis, D. A. Power, A. T. Sim, and B. E. Kemp, Coordinated control of endothelial nitric-oxide synthase phosphorylation by protein kinase C and the cAMP-dependent protein kinase, J Biol Chem 276 (2001) 17625-17628. [15] B. J. Michell, M. B. Harris, Z. P. Che n, H. Ju, V. J. Venema, M. A. Blackstone, W. Huang, R. C. Venema, and B. E. Kemp, Identification of re gulatory sites of phosphorylation of the bovine e ndothelial nitric-oxide synthase at serine 617 and serine 635, J Biol Chem 277 (2002) 42344-42351. [16] S. W. Bae, H. S. Kim, Y. N. Cha, Y. S. Park, S. A. J o, and I. Jo, Rapid increase in endothelial nitric oxide pr oduction by bradykinin is medi ated by protein kinase A signaling pathway, Bioche m Biophys Res Commun 306 (2003) 981-987. [17] C. Rask-Madsen, and G. L. King, Di fferential regulation of VEGF signaling by PKC-alpha and PKC-epsilon in endothelial ce lls, Arterioscler Thromb Vasc Biol 28 (2008) 919-924. [18] C. Partovian, Z. Zhuang, K. Moodie, M. Lin, N. Ouchi, W. C. Sessa, K. Walsh, and M. Simons, PKCalpha activates eNOS and incr eases arterial blood flow in vivo, Circ Res 97 (2005) 482-487.

PAGE 176

158 [19] D. Fulton, J. P. Gratton, T. J. McCabe, J. Fontana, Y. Fujio, K. Walsh, T. F. Franke, A. Papapetropoulos, and W. C. Sessa, Regula tion of endothelium-derived nitric oxide production by the protein kinase Akt, Nature 399 (1999) 597-601. [20] W. Xi, H. Satoh, H. Kase, K. Suzuki and Y. Hattori, Stimulated HSP90 binding to eNOS and activation of the PI3-Akt pathway contribute to globular adiponectin-induced NO production: vasorelaxation in response to globular adiponectin, Biochem Biophys Res Commun 332 (2005) 200-205. [21] G. Zeng, F. H. Nystrom, L. V. Ravicha ndran, L. N. Cong, M. Kirby, H. Mostowski, and M. J. Quon, Roles for insulin receptor, PI3-kinase, and Akt in insulin-signaling pathways related to production of nitric oxide in human vascul ar endothelial cells, Circulation 101 (2000) 1539-1545. [22] H. Cai, Z. Li, M. E. Davis, W. Kanne r, D. G. Harrison, and S. C. Dudley, Jr., Aktdependent phosphorylation of serine 1179 and mitogen-ac tivated protein kinase kinase/extracellular sign al-regulated kinase 1/2 cooperatively medi ate activation of the endothelial nitric-oxide synthase by hydrogen peroxide, Mol Pharmacol 63 (2003) 325331. [23] J. C. Mason, R. Steinberg, E. A. Lidington, A. R. Kinderlerer, M. Ohba, and D. O. Haskard, Decay-accelerating factor induction on vascular endothelium by vascular endothelial growth factor (VEGF) is medi ated via a VEGF receptor-2 (VEGF-R2)and protein kinase C-alpha/epsil on (PKCalpha/epsilon)-dependent cytoprotective signaling pathway and is inhibited by cy closporin A, J Biol Chem 279 (2004) 41611-41618. [24] H. Li, S. A. Oehrlein, T. Wallerath, I. Ihrig-Biedert, P. Wohlfart, T. Ulshofer, T. Jessen, T. Herget, U. Forstermann, and H. Klei nert, Activation of protein kinase C alpha and/or epsilon enhances tran scription of the human endothe lial nitric oxide synthase gene, Mol Pharmacol 53 (1998) 630-637. [25] J. Li, X. Hu, P. Selvakumar, R. R. Ru ssell, 3rd, S. W. Cushman, G. D. Holman, and L. H. Young, Role of the nitric oxide pa thway in AMPK-mediated glucose uptake and GLUT4 translocation in heart muscle Am J Physiol Endocrinol Metab 287 (2004) E834841. [26] V. A. Morrow, F. Foufelle J. M. Connell, J. R. Petrie, G. W. Gould, and I. P. Salt, Direct activation of AMP-activat ed protein kinase stimulates nitric-oxide synthesis in human aortic endothelial cells, J Biol Chem 278 (2003) 31629-31639. [27] Z. P. Chen, K. I. Mitchelh ill, B. J. Michell, D. Staple ton, I. Rodriguez-Crespo, L. A. Witters, D. A. Power, P. R. Ortiz de Mont ellano, and B. E. Kemp, AMP-activated protein kinase phosphorylation of endothelial NO synthase, FEBS Lett 443 (1999) 285-289.

PAGE 177

159 [28] J. A. Reihill, M. A. Ewart, D. G. Hardie, and I. P. Salt, AMP-activated protein kinase mediates VEGF-stimulated endot helial NO production, Biochem Biophys Res Commun 354 (2007) 1084-1088. [29] Y. Zhang, T. S. Lee, E. M. Kolb, K. Sun, X. Lu, F. M. Sladek, G. S. Kassab, T. Garland, Jr., and J. Y. Shyy, AMP-activated pr otein kinase is involve d in endothelial NO synthase activation in response to shear st ress, Arterioscler Thromb Vasc Biol 26 (2006) 1281-1287. [30] B. J. Davis, Z. Xie, B. Viollet, a nd M. H. Zou, Activation of the AMP-activated kinase by antidiabetes drug metformin stimul ates nitric oxide synthesis in vivo by promoting the association of heat shock protein 90 and endothelial n itric oxide synthase, Diabetes 55 (2006) 496-505. [31] Y. Ido, D. Carling, and N. Ruderman, Hyperglycemia-induced apoptosis in human umbilical vein endothelial cells: inhibition by the AMP-activated protein kinase activation, Diabetes 51 (2002) 159-167. [32] L. G. Fryer, A. ParbuPatel, and D. Carling, The Anti-diabetic drugs rosiglitazone and metformin stimulate AMP-activated protein kinase through distinct signaling pathways, J Biol Chem 277 (2002) 25226-25232. [33] C. Blazquez, M. J. Geelen, G. Velasc o, and M. Guzman, The AMP-activated protein kinase prevents ceramide synthesis de novo and apoptosis in astrocytes, FEBS Lett 489 (2001) 149-153. [34] I. Fleming, C. Schulz, B. Fichtlscherer, B. E. Kemp, B. Fisslthaler, and R. Busse, AMP-activated protein kinase (AMPK) regul ates the insulin-induced activation of the nitric oxide synthase in huma n platelets, Thromb Haemost 90 (2003) 863-871. [35] T. Okayasu, A. Tomizawa, K. Suzuki K. Manaka, and Y. Hattori, PPARalpha activators upregulate eNOS activity and inhi bit cytokine-induced NF-kappaB activation through AMP-activated protein ki nase activation, Life Sci 82 (2008) 884-891. [36] S. Frame, and D. Zheleva, Targeti ng glycogen synthase kinase-3 in insulin signalling, Expert Opin Ther Targets 10 (2006) 429-444. [37] G. Y. Oudit, H. Sun, B. G. Kerfant, M. A. Crackower, J. M. Penninger, and P. H. Backx, The role of phosphoinositide-3 kinase and PTEN in cardiovascular physiology and disease, J Mol Cell Cardiol 37 (2004) 449-471. [38] A. M. Al-Khouri, Y. Ma, S. H. Togo, S. Williams, and T. Mustelin, Cooperative phosphorylation of the tumor suppressor phos phatase and tensin homologue (PTEN) by casein kinases and glycogen synthase kinase 3beta, J Biol Chem 280 (2005) 3519535202.

PAGE 178

160 [39] H. S. Kim, C. Skurk, S. R. Thomas, A. Bialik, T. Suhara, Y. Kureishi, M. Birnbaum, J. F. Keaney, Jr., and K. Walsh, Regulation of angiogenesis by glycogen synthase kinase3beta, J Biol Chem 277 (2002) 41888-41896. [40] D. M. Greif, D. B. Sacks, a nd T. Michel, Calmodulin phosphorylation and modulation of endothelial nitric oxide synthase catalysis, Proc Natl Acad Sci U S A 101 (2004) 1165-1170. [41] K. Cieslik, C. M. Lee, J. L. Tang, and K. K. Wu, Transcri ptional regulation of endothelial nitric-oxide syntha se by an interaction between casein kinase 2 and protein phosphatase 2A, J Biol Chem 274 (1999) 34669-34675. [42] Y. Miyata, and E. Nishida, CK2 binds phosphorylates, and regulates its pivotal substrate Cdc37, an Hsp90-cochaperone, Mol Cell Biochem 274 (2005) 171-179. [43] S. Kosuge, Y. Sawano, and K. Ohtsuki, A novel CK2-mediated activation of type II cAMP-dependent protein kinase through specific phosphorylation of its regulatory subunit (RIIalpha) in vitro, Biochem Biophys Res Commun 310 (2003) 163-168. [44] D. D. Williams, O. Marin, L. A. Pinna and C. G. Proud, Phosphorylated seryl and threonyl, but not tyrosyl, re sidues are efficient specificity determinants for GSK-3beta and Shaggy, FEBS Lett 448 (1999) 86-90. [45] O. Marin, V. H. Bustos, L. Cesaro, F. Meggio, M. A. Pagano, M. Antonelli, C. C. Allende, L. A. Pinna, and J. E. Allend e, A noncanonical sequence phosphorylated by casein kinase 1 in beta-catenin may play a ro le in casein kinase 1 targeting of important signaling proteins, Proc Natl Acad Sci U S A 100 (2003) 10193-10200. [46] G. Huang, S. Chen, S. Li, J. Cha, C. Long, L. Li, Q. He, and Y. Liu, Protein kinase A and casein kinases mediate sequential phosphor ylation events in th e circadian negative feedback loop, Genes Dev 21 (2007) 3283-3295. [47] M. Zanetti, R. Barazzoni, M. Stebel, E. R oder, G. Biolo, F. E. Baralle, L. Cattin, and G. Guarnieri, Dysregulation of the endotheli al nitric oxide synthase-soluble guanylate cyclase pathway is normalized by insulin in the aorta of diabetic rat, Atherosclerosis 181 (2005) 69-73. [48] Z. Xu, X. Ji, and P. G. Boysen, Exoge nous nitric oxide generates ROS and induces cardioprotection: involvement of PKG, mitochondrial KATP channels, and ERK, Am J Physiol Heart Circ Physiol 286 (2004) H1433-1440. [49] C. Chen, V. A. Korshunov, M. P. Masse tt, C. Yan, and B. C. Berk, Impaired vasorelaxation in inbred mice is associated w ith alterations in both nitric oxide and super oxide pathways, J Vasc Res 44 (2007) 504-512.

PAGE 179

161 [50] T. Yamashita, S. Kawashima, Y. Ohashi, M. Ozaki, Y. Rikitake, N. Inoue, K. Hirata, H. Akita, and M. Yokoyama, Mechanisms of reduced nitric oxide/cGMP-mediated vasorelaxation in transgenic mice overexpre ssing endothelial nitr ic oxide synthase, Hypertension 36 (2000) 97-102. [51] S. Borniquel, I. Valle, S. Cadenas, S. Lamas, and M. Monsalve, Nitric oxide regulates mitochondrial oxidative stress prot ection via the transcri ptional coactivator PGC-1alpha, Faseb J 20 (2006) 1889-1891. [52] M. Cudmore, S. Ahmad, B. Al-Ani, P. Hewett, S. Ahmed, and A. Ahmed, VEGF-E activates endothelial nitric oxide synthase to induce a ngiogenesis via cGMP and PKGindependent pathways, Bi ochem Biophys Res Commun 345 (2006) 1275-1282. [53] T. A. John, B. O. Ibe, and J. U. Raj, Regulation of endothelial nitric oxide synthase: involvement of protein kinase G 1 beta, se rine 116 phosphorylation and lipid structures, Clin Exp Pharmacol Physiol 35 (2008) 148-158. [54] A. E. Linder, L. P. McCluskey, K. R. Cole, 3rd, K. M. Lanning, and R. C. Webb, Dynamic association of nitric oxide downs tream signaling molecules with endothelial caveolin-1 in rat aorta, J Pharmacol Exp Ther 314 (2005) 9-15. [55] B. R. Flam, P. J. Hartmann, M. Harrell -Booth, L. P. Solomonson, and D. C. Eichler, Caveolar localization of argi nine regeneration enzymes, ar gininosuccinate synthase, and lyase, with endothelial nitric oxide synthase, Nitric Oxide 5 (2001) 187-197. [56] E. Butt, M. Bernhardt, A. Smolenski, P. Kotsonis, L. G. Frohlich, A. Sickmann, H. E. Meyer, S. M. Lohmann, and H. H. Schmid t, Endothelial nitricoxide synthase (type III) is activated and becomes calcium independent upon phosphorylation by cyclic nucleotide-dependent protein kinases, J Biol Chem 275 (2000) 5179-5187. [57] D. Feliers, X. Chen, N. Akis, G. G. Choudhury, M. Madaio, and B. S. Kasinath, VEGF regulation of endothelial nitric oxide synthase in glomerular endothelial cells, Kidney Int 68 (2005) 1648-1659. [58] S. Kliche, and J. Waltenberger, VEGF receptor signaling and endothelial function, IUBMB Life 52 (2001) 61-66. [59] J. Igarashi, P. A. Erwin, A. P. Dantas, H. Chen, and T. Michel, VEGF induces S1P1 receptors in endothelial cells: Implications for cross-talk between sphingolipid and growth factor receptors, Proc Natl Acad Sci U S A 100 (2003) 10664-10669. [60] T. Tanimoto, Z. G. Jin, and B. C. Be rk, Transactivation of vascular endothelial growth factor (VEGF) receptor Flk-1/KDR is involved in sphingosine 1-phosphatestimulated phosphorylation of Akt and endotheli al nitric-oxide synthase (eNOS), J Biol Chem 277 (2002) 42997-43001.

PAGE 180

162 [61] A. Brouet, P. Sonveaux, C. Dessy, J. L. Balligand, and O. Feron, Hsp90 ensures the transition from the early Ca2+-dependent to the late phosphorylation-dependent activation of the endothelial nitric-oxide syntha se in vascular endothe lial growth factorexposed endothelial cells, J Biol Chem 276 (2001) 32663-32669. [62] S. Takahashi, and M. E. Mendelsohn, Synergistic activation of endothelial nitricoxide synthase (eNOS) by HSP90 and Ak t: calcium-independent eNOS activation involves formation of an HSP90-Akt-CaM-bound eNOS complex, J Biol Chem 278 (2003) 30821-30827. [63] S. Jesmin, S. Zaedi, N. Shimojo, M. Iemitsu, K. Masuzawa, N. Yamaguchi, C. N. Mowa, S. Maeda, Y. Hattori, and T. Miya uchi, Endothelin antagonism normalizes VEGF signaling and cardiac function in STZ-induced diabetic rat hearts, Am J Physiol Endocrinol Metab 292 (2007) E1030-1040. [64] T. Kobayashi, and K. Kamata, Short-term insulin treatment and aortic expressions of IGF-1 receptor and VEGF mRNA in diabetic ra ts, Am J Physiol Heart Circ Physiol 283 (2002) H1761-1768. [65] L. J. Shen, K. Beloussow, and W. C. Shen, Accessibility of endothelial and inducible nitric oxide synthase to the intracellula r citrulline-arginine regeneration pathway, Biochem Pharmacol 69 (2005) 97-104. [66] Y. Tsurumi, T. Murohara, K. Krasinski, D. Chen, B. Witzenbich ler, M. Kearney, T. Couffinhal, and J. M. Isner, Reciprocal re lation between VEGF a nd NO in the regulation of endothelial integrity, Nat Med 3 (1997) 879-886. [67] R. Govers, and T. J. Rabelink, Cellular regulation of endothelial nitric oxide synthase, Am J Physiol Renal Physiol 280 (2001) F193-206. [68] R. C. Venema, Post-translational mechan isms of endothelial nitric oxide synthase regulation by bradykinin, Int Immunopharmacol 2 (2002) 1755-1762. [69] D. M. Dudzinski, and T. Michel, Life history of eNOS: partners and pathways, Cardiovasc Res 75 (2007) 247-260. [70] S. Stratford, K. L. Hoehn, F. Liu, and S. A. Summers, Regulation of insulin action by ceramide: dual mechanisms linking ceramide accumulation to the inhibition of Akt/protein kinase B, J Biol Chem 279 (2004) 36608-36615. [71] H. Li, P. Junk, A. Huwiler, C. Burkhard t, T. Wallerath, J. Pfeilschifter, and U. Forstermann, Dual effect of ceramide on huma n endothelial cells: induction of oxidative stress and transcriptional upr egulation of endothelial nitric oxide synthase, Circulation 106 (2002) 2250-2256.

PAGE 181

163 [72] P. Der, J. Cui, and D. K. Das, Role of lipid rafts in ceramide and nitric oxide signaling in the ischemic and preconditioned hearts, J Mol Cell Cardiol 40 (2006) 313320. [73] S. Morandell, T. Stasyk, K. Grosstessner-Hain, E. Roitinger, K. Mechtler, G. K. Bonn, and L. A. Huber, Phosphoproteomics strate gies for the functional analysis of signal transduction, Proteomics 6 (2006) 4047-4056. [74] Z. Songyang, K. P. Lu, Y. T. Kwon, L. H. Tsai, O. Filhol, C. Cochet, D. A. Brickey, T. R. Soderling, C. Bartleson, D. J. Graves, A. J. DeMaggio, M. F. Hoekstra, J. Blenis, T. Hunter, and L. C. Cantley, A st ructural basis for substrate sp ecificities of protein Ser/Thr kinases: primary sequence preference of casein kinases I and II, NIMA, phosphorylase kinase, calmodulin-dependent kinase II, CDK5, and Erk1, Mol Cell Biol 16 (1996) 64866493. [75] R. B. Pearson, and B. E. Kemp, Prot ein kinase phosphorylation site sequences and consensus specificity motifs: tabulations, Methods Enzymol 200 (1991) 62-81. [76] J. Li, X. Zhao, X. Li, K. M. Lerea, a nd S. C. Olson, Angiotensin II type 2 receptordependent increases in nitric oxide synthase expression in the pulmonary endothelium is mediated via a G alpha i3/Ras/Raf/MAPK pathway, Am J Physiol Cell Physiol 292 (2007) C2185-2196. [77] A. Ptasinska, S. Wang, J. Zhang, R. A. Wesley, and R. L. Danner, Nitric oxide activation of peroxisome proliferatoractivated receptor gamma through a p38 MAPK signaling pathway, Faseb J 21 (2007) 950-961. [78] E. Anter, S. R. Thomas, E. Schulz, O. M. Shapira, J. A. Vita, and J. F. Keaney, Jr., Activation of endothelial nitric-oxide synthase by the p38 MAPK in response to black tea polyphenols, J Biol Chem 279 (2004) 46637-46643. [79] Z. Yuan, W. Feng, J. Hong, Q. Zhe ng, J. Shuai, and Y. Ge, p38MAPK and ERK promote nitric oxide production in cultured human retinal pigmented epithelial cells induced by high concentrati on glucose, Nitric Oxide (2008). [80] P. A. Erwin, A. J. Lin, D. E. Golan, and T. Michel, Receptor-regulated dynamic Snitrosylation of endothelial nitric-oxide synt hase in vascular endot helial cells, J Biol Chem 280 (2005) 19888-19894. [81] F. A. Sanchez, N. B. Savalia, R. G. Dura n, B. K. Lal, M. P. Boric, and W. N. Duran, Functional significance of di fferential eNOS translocation, Am J Physiol Heart Circ Physiol 291 (2006) H1058-1064.

PAGE 182

164 [82] B. Musicki, M. F. Kramer, R. E. B ecker, and A. L. Burnett, Inactivation of phosphorylated endothelial nitric oxide syntha se (Ser-1177) by O-Gl cNAc in diabetesassociated erectile dysfunction, Proc Natl Acad Sci U S A 102 (2005) 11870-11875. [83] A. Nandi, R. Sprung, D. K. Barma, Y. Zhao, S. C. Kim, J. R. Falck, and Y. Zhao, Global identification of O-GlcNAc-m odified proteins, Anal Chem 78 (2006) 452-458.

PAGE 183

165 CHAPTER THREE ARGININOSUCCINATE SYNTHASE SUBCELLULAR LOCALIZATION AND PROTEIN INTERACTIONS Overview Argininosuccinate synthase (A S) is an essential mediator of endothelial health by providing a dedicated source of arginine for nitric oxide (NO) production and promoting endothelial cell viability. Our laboratory has previously demonstrated that AS is present in endothelial caveolar fractions along with endothelial nitr ic oxide synthase (eNOS) and argininosuccinate lyase (AL), the core co mponents of the citrulline-NO cycle. The studies in this Chapter were designed to de fine the subcellular localization of AS in relationship to these components, to determine whether ther e are interactions that are essential for nitric oxide metabolism and to begin characterizing additional novel components of the nitric oxide metabolom e. Utilizing immunofluorescence microscopy, we found that AS localizes to perinuclear regions and the plasma membrane. Further, eNOS localization overlaps AS localization suggesting functional re levance We also found that AS colocalizes with caveolin-1, a key regulator of eNOS function, in distinct plasma membrane regions and the Golgi. Utilizing co-immunoprecipitation studies, we found that AS interacts with HSP90 and caveolin-1. These two proteins are intimately

PAGE 184

166 involved in the regulation of endothelial nitric oxide producti on. We determined that AS and co-localizes with HSP90 utilizing immunofluorescence. We also identified a caveolin binding motif in the AS protein sequence that suggests a direct in teraction. Finally, to begin characterizing the nitric oxide metabol ome from a more global perspective, we utilized co-immunoprecipitations followed by mass spectrometry. We identified several putative interacting partners that are either novel or understudied in the regulation of NO production. Overall, our work defines a tightly coupled system for the regulation of nitric oxide production and highlights the intricat e and dynamic nature of citrulline-NO cycle localization and interactions. Materials and Methods Immunofluorescence: BAEC were plated out on chamber slides. Cells were fixed with 3.7% paraformaldehyde and permeb ealized with 0.05% triton X-100. After blocking, cells were incubated with the follo wing antibodies: AS (Everest Biotech), eNOS, HSP90 and caveolin-1 (BD Bioscience s). Cells were then incubated with fluorescently labeled secondary antibodies (Invitrogen). Images were generated with a Nikon Eclipse E1000 Fluorescent Microscope running Genus 2.81 software from Applied Imaging. On all experiments, negative cont rols were conducted by staining one chamber with secondary antibodies only. Immunoprecipitation and protein identification using LC-MS/MS: Coimmunoprecipitation studies were conducted with the following antibodies: AS (Everest

PAGE 185

167 Biotech), HSP90, caveolin-1, eNOS (BD Biosciences), calmodulin I, NOSIP, PKG, AL (Santa Cruz), AMPK PKC PKAc CKII, and Akt (Cell Signaling Technology) utilizing the methods described in Chapter Two, Page 101. For samples that were utilized for mass spec, processing occurred very rapidly due to the labile nature of post-translati onal modifications and some protein-protein interactions. In addition, RIPA buffer was utilized as the lysis and wash buffer since it is more stringent and diminishes the possibility of non-specific interac tions. Eluted protein complexes were subject to SDS-PAGE, bands of interest were excised, then digested proteins were subjected to liquid chro matography-tandem mass spectrometry as described in Chapter Two, Page 108. Bioinformatics: The AS caveolin binding motif wa s identified utilizing Expasy Prosite (Scan Prosite Tool: ht tp://www.expasy.ch/tools/scanpr osite/) by searching for the following motif in the AS protein sequence: [WFY] X X X X [WFY] X X [WFY]. The motif was then identified on th e 3D crystal structure of AS utilizing ViewerLite software version 5.0 (Accelrys Corporation, San Diego, California). Results AS Subcellular Localization Overlaps with eNOS and Caveolin-1: The subcellular localization of endothelial AS has not previously been charact erized and we believe that it is a key regulatory feature of the citrulline-NO cycle. We hypothesized that for optimal

PAGE 186

168 function of the citrulline-NO cycle, AS and eNOS should localize to the same regions within the cell. Utilizing i mmunofluorescence microscopy, we demonstrated that AS and eNOS colocalize and distribute in endothelial cells in a sim ilar fashion, predominantly in the Golgi and the cytoplasmic membrane (Fig ure 27). We have previously demonstrated that caveolin-1 and AS co-fractionate in crude caveolar cell extr acts, but we did not determine whether these two proteins were in the same or independent caveolar regions. Since eNOS and caveolin-1 colocalize and physi cally interact, we hypot hesized that there would be distinct regions of overlap in th e localization of AS and caveolin-1. As shown in Figure 25, AS co-localizes with caveolin -1 in distinct membrane regions and the Golgi, much like what is seen with eNOS.

PAGE 187

Figure 27: AS Colocalizes wi th eNOS and Caveolin-1. Immunofluorescence microscopy images demonstrating the loca lization of AS in the Golgi and plasma membrane along with its colocalization with eNOS (A) and caveolin-1 (B). AS is red, eNOS or caveolin are green and the merged image shows regions of co-localization in yellow. The nucleus is stained in blue. Arrows point to selected areas of colocalization in the Golgi and plasma membrane (n = 3). (C) Secondary only negative control. A AS eNOS Merge B AS Caveolin Merge C 169

PAGE 188

AS Protein Interactions: In order to begin deciphering key AS interacting partners, a series of immunoprecipitations (IP) were carried out. When utilizing AS as the IP antibody, two interactions were consistently noted: HSP 90 and caveolin-1 (Figure 28). These interactions were ve rified by immunoprecipitaing with HSP90 and caveolin-1 then probing with AS (Figure 28). Figure 28: AS Co-Immunoprecipitate s with HSP90 and Caveolin-1. (A) Representative blot (n = 3) of BAEC lysates that were immunoprecipitated with an AS antibody (IP AS). The membrane was probed with HSP90 (IB HSP90) and caveolin-1 (IB Cav-1). (B) Representative blot (n = 2) of BAEC lysates that were immunoprecipitated with an HSP90 antibody (IP HSP90). The membrane was probed with AS (IB AS). (C) Representative blot (n = 2) of BAEC lysates that were immunoprecipitated with a caveolin-1 antibody (IP Cav-1). The membrane was probed with AS (IB AS). IP HSP90 IB AS B IP AS IB HSP90 IB Cav-1 A IP Cav-1 IB AS C 170

PAGE 189

These results prompted us to assess the subcellular localization of AS and HSP90. As shown in Figure 29, there appears to be co-localization between AS and HSP90, although the specific regions are difficult to discern since HSP 90 is widely distributed in the cytoplasm. Figure 29: AS Colocalizes with HSP90. (A) Immunofluorescence microscopy images demonstrating colocalization of AS with HSP90 (n = 1). AS is red, HSP90 is green and the merged image shows regions of co-localization in yellow. The nucleus is stained in blue. (B) Secondary only negative control. AS HSP90 Merge A B 171

PAGE 190

Since eNOS has a caveolin binding motif, we decided to search for the same motif in the AS protein sequence. We identified the sequence 317-FAELVYTGF-325 which fits the pattern of the caveolin binding motif found in eNOS: [WFY]-X-X-X-X[WFY]-X-X-[WFY] (Figure 30). This implie s a direct interac tion between AS and caveolin. Figure 30: AS has a Caveolin Binding Motif. Three dimensional structure of Human AS (PDB ID 2NZ2) demonstrating the region containing the following caveolin binding motif: [WFY]-X-X-X-X-[WFY]-X-X-[WFY]. The motif corresponds to the sequence: 317-FAELVYTGF-325. Identified phosphorylation sites and substrates are also shown as described in Figure 22, Chapter Two, Page 135. Citrulline Caveolin binding motif S180 S189 Aspartate T134 S131 S328 172

PAGE 191

173 It is also important to note that the in vestigation of AS prot ein interactions via immunoprecipitation was quite comprehensive and many other proteins were explored. We did obtain some positive preliminary results with AS interactions with eNOS, AL, PKA, PKC and calmodulin. Taken together, our re sults demonstrate that there are key protein interactions that potentially mediate AS function and the function of the citrulline-NO cycle as a whole. Proteomic Examination of th e Nitric Oxide Metabolome: In order to begin characterizing the NO metabolome, we util ized immunoprecipitation with an AS or eNOS antibody followed by LC-MS/MS. BAEC cell lysates were collected in RIPA buffer and immunoprecipitation was conducted. Putative interacting components were separated by SDS-PAGE then specific bands were cut from the gel and subjected to mass spectrometry for protein identification. This revealed several interesting interacting partners that are relevant to known regulatory mechanisms of NO production. A list of the most relevant interactions and their know n or potential roles in NO biology is shown in Table 8. Some interactions were specific fo r AS while others were identified with both AS and eNOS antibodies. Interestingly, AS and eNOS did not seem to interact with each other with this methodology, a lthough the interaction has been noted previously with other methodologies (Brenda Fl am, unpublished results). Anot her interaction that was missing from the list was the well document ed eNOS-HSP90 interaction. The AS-HSP90 interaction was confirmed with the mass spec data. Table 9 s hows all 85 proteins found to co-IP with either AS or eNOS.

PAGE 192

Table 8: The Basal Nitr ic Oxide Metabolome. Table showing a few of the interactions identified utilizing basal cell ly sates that were immunoprecipitated with an AS antibody. After separation via SDS-PAGE, bands were excised, digested and subjected to liquid chromatography-tandem mass spectrometry. Table also shows role of these interacting partners in regulating NO metabolism (n = 1). PROTEIN FUNCTION IP ANTIBODY cGMP-dependent protein kinase Kinase activated by soluble guanylylcyclase. Important for vasodilation and regulates eNOS. AS eNOS Dynamin Large GTP-binding protein residing within similar membrane compartments as eNOS. Interacts with eNOS and increases its activity. AS Golgi SNAP Receptor Complex Involved in transport from the ER to the Golgi apparatus as well as in intra-Golgi transport. Disruption of intracellular trafficking has been associated with cardiovascular disorders. AS eNOS HSP90 Interacts with eNOS and increases its activity. AS Kininogen-1 precursor Cleaved into several products, including bradykinin, which is an important regulator of vasodilation. eNOS is known to interact with the bradykinin receptor. This interaction is inhibitory and is released upon treatment with bradykinin. AS eNOS Prohibitin Mitochondrial protein that protects endothelial cells from reactive-oxygen species damage. AS eNOS 174

PAGE 193

Table 9: Putative AS and eNOS Interacting Partners. Complete list of all proteins identified via IP with an AS and/or eNOS antibody followed by mass spec as described in Table 8 above. Actin, cytoplasmic 2 Bos taurus (Bovine) Trypsin precursor Sus scrofa (Pig) Tropomyosin alpha-1 chain Bos taurus (Bovine) Myosin-10 Bos taurus (Bovine) Myosin regulatory light chain 2, smooth muscle isoform Bos taurus (Bovine) Myosin-9 Canis familiaris (Dog) Actin-like protein 3 Bos taurus (Bovine) Histone H2A type 1 Bos taurus (Bovine) Ornithine decarboxylase antizyme Bos taurus (Bovine) Cationic trypsin precursor Bos taurus (Bovine) Actin, alpha skeletal muscle Bos taurus (Bovine) Myosin light polypeptide 6 Bos taurus (Bovine) ADP/ATP translocase 2 Tachyglossus acu leatus aculeatus (Australian echidna) Vimentin Bos taurus (Bovine) Kininogen-1 precursor Bos taurus (Bovine) Ribonuclease pancreatic Hippopotam us amphibius (Hippopotamus) Histone H2A type 2-C Bos taurus (Bovine) Ribonuclease pancreatic Antilocapra americana (Pronghorn) Dynamin-1-like protein Bos taurus (Bovine) Ubiquitin Bos taurus (Bovine) 130 kDa phosphatidylinositol 4,5-biphosphate-dependent ARF1 GTPase-activating protein Bos taurus (Bovine) Actin-related protein 2/3 complex subunit 4 Bos taurus (Bovine) Myosin-Id Bos taurus (Bovine) Heterogeneous nuclear ribonucleoprot ein A1 Bos taurus (Bovine) Actin-related protein 2/3 complex subunit 3 Bos taurus (Bovine) Serum albumin precursor Bos taurus (Bovine) Histone H2B type 1-K Bos taurus (Bovine) Golgi SNAP receptor complex member 1 Bos taurus (Bovine) Alpha-S1-casein precursor Bos taurus (Bovine) Glycoprotein GIII precursor Bovine herpesvirus 1.1 (strain Cooper) (BoHV-1) (Infectious bovine rhinotracheitis virus) Prohibitin-2 Bos taurus (Bovine) Abnormal spindle-like microcephaly-associated protein homolog Ovis aries (Sheep) Alpha-enolase Bos taurus (Bovine) Actin-related protein 2/3 complex subunit 1B Bos taurus (Bovine) 60S ribosomal protein L4 Bos taurus (Bovine) Junction plakoglobin Bos taurus (Bovine) Polymeric-immunoglobulin receptor precursor Bos taurus (Bovine) Tubulin alpha-1A chain Sus scrofa (Pig) NADH dehydrogenase [ubiquinone] 1 alpha subc omplex subunit 12 Bos taurus (Bovine) Pyruvate kinase isozyme M1 Felis silvestris catus (Cat) von Willebrand factor precursor Canis familiaris (Dog) Tyrosine-protein kinase SYK Sus scrofa (Pig) 175

PAGE 194

176 Table 9 (continued) Endoplasmin precursor Bos taurus (Bovine) Gamma-aminobutyric-acid receptor subunit alpha-3 precursor Bos taurus (Bovine) Elongation factor 1-alpha 1 Bos taurus (Bovine) Histone H3.1 Bos taurus (Bovine) 60S acidic ribosomal protein P0 Bos taurus (Bovine) cGMP-dependent protein kinase 1, alpha isozyme Bos taurus (Bovine) Calpain-1 catalytic subunit Oryctolagus cuniculus (Rabbit) Beta-casein precursor [Contains: Ca soparan] Bos taurus (Bovine) Ribosome recycling factor, mitochondrial precursor Bos taurus (Bovine) Heterogeneous nuclear ribonucleoprotei ns A2/B1 Bos taurus (Bovine) Rho guanine nucleotide exchange factor 9 Bos taurus (Bovine) 52 kDa Ro protein Bos taurus (Bovine) Telomerase reverse transcriptase Canis familiaris (Dog) Fatty acid-binding protein, adipocyte Sus scrofa (Pig) Nitric oxide synthase, induc ible Bos taurus (Bovine) Desmoglein-1 precursor Canis familiaris (Dog) Heat shock 70 kDa protein 4 Canis familiaris (Dog) Alpha-S2-casein precursor [Contains: Casocidin-1 Bos taurus (Bovine) Serpin H1 precursor Bos taurus (Bovine) Coiled-coil domain-containing protein 113 Bos taurus (Bovine) Cofilin-2 Bos taurus (Bovine) Histone H1.1 Bos taurus (Bovine) Pro-epidermal growth factor precursor Canis familiaris (Dog) Glial fibrillary acidic protein Bos taurus (Bovine) Probable phospholipid-transporting ATPase IF Oryctolagus cuniculus (Rabbit) Histone H3.2 Bos taurus (Bovine) Heat shock protein HSP 90-al pha Bos taurus (Bovine) 5-hydroxytryptamine 1A recep tor Equus caballus (Horse) Tripartite motif-containing prot ein 9 Bos taurus (Bovine) Cystic fibrosis transmembrane conductance regulator Sus scrofa (Pig) An additional piece of data related to protein interactions was uncovered when searching for AS phosphorylation sites via mass spec. When AS was overexpressed and then treated with okadaic acid, T131 was identified as a phosphoryl ation site (Chapter Two, Figure 19, Page 129). In addition, several other bands were cut out from the gel that contained purified AS and any proteins that might have come down with it. One notable protein identified was NOSIP (eNOS inte racting protein). NOSIP is a known binding partner with eNOS and is important for its subcellular localization. NOSIP promotes

PAGE 195

177 translocation of eNOS from the plasma membrane to intracellular sites, thereby uncoupling eNOS from caveolae and inhibiting NO synthesis. Discussion In this Chapter, the subcellular localiza tion of endothelial AS was defined, its colocalization with eNOS, caveolin-1 and HSP90 was described and important AS interacting partners were identified. This is the second post-tr anslational regulatory mechanism addressed in this dissertation. The subcellular localization of AS ha s been defined in several tissues. For instance, in untransfected or AS-transfected VSMC, AS is de tected in both cytosolic and membrane fractions via western blot [1]. AS expression is increased in membrane fractions upon stimulation with LPC/IFN suggesting that AS is transported to membrane regions for functionally rele vant purposes [1, 2]. In addition, immunohistochemical studies of AS-transfect ed cells indicates a punctuate pattern of expression that suggests mitochondrial localizati on [1], much like what is seen in liver [2]. This study supported the concept of substrate channeling [1], which would necessitate co-localization and assembly into a functional complex. Our identification of AS localization to the peri nuclear/Golgi region and plas ma membrane supports those previous findings since NO is produced in these two subcellular compartments [3-5]. Thus, much like urea cycle enzymes, the ci trulline-NO cycle enzymes form functional

PAGE 196

178 complexes in specific cellular compartments th at allow efficient channeling of substrates so that the arginine required for NO production is distinct fr om bulk cellular arginine. The distinct localization pa ttern for AS in endothelia l cells also led to an important observation. It is possible that AS has localization-specific or organellespecific functions. These functi ons might be related to post -translational modifications. Therefore, some of the difficulties in obtaining definitive mass spectrometry data may be related to the fact that we utilized whole cell lysates to look for AS phosphorylation sites. This may in fact have led to a dilution of the data since perhaps certain regions of the cells have a high population of phosphorylated AS while othe rs, perhaps most, dont. It will be important to determine the phosphor ylation pattern in different cellular compartments to gain a better understanding of the impact and regulation of these modifications. eNOS is also known to localize to the Go lgi and plasma membrane [5] and we did find that AS and eNOS co-localize. Alt hough it is unclear how AS functions within specific cellular compartments, there are two distinct pools of active eNOS, one in the Golgi and one in the plasma membrane. Th e relative importance of these two pools of eNOS is still under investigation, but one study demonstrated that although NO is produced each of these regions, the plasma membrane produces significantly more NO [3]. In that study, they generated endothelial cells that expressed wild type eNOS, Golgionly eNOS or plasma membrane-only eNOS without altering any other components of the citrulline-NO cycle. This suggested that e ither the other enzymes of the cycle are also

PAGE 197

179 present in Golgi and plasma membrane or th at there is another source of substrate for eNOS. Our studies and those of others suggest that endothelial cells are dependent on recycling of citrulline to argi nine as a source of eNOS substrate [6, 7]. The fact that AS localized to those regions s upports those studies. It would be interesting to design a similar study where both AS and eNOS loca lization could be manipulated. Then, the dependence of co-localization could be definitively proven. However, the studies with Golgi versus plasma membrane eNOS were done by first knocking out endogenous eNOS and then re-constituting the endothelial cells with the different eNOS constructs [3]. We have shown that moderate AS knockdown leads to significant endothelial apoptosis [8], so the same approach might not be successful for AS. This suggests that while the function of eNOS ma y be exclusively to produce NO, AS might have several different functions, including as anti-apoptot ic signaling [8], that make it even more important for overall vascular biology. In addition, we demonstrated that AS lo calized in distinct plasma membrane regions in conjunction with caveolin-1 [7, 9]. This expa nds our previous work by demonstrating that the regions where AS lo calizes in the plasma membrane are also regions where caveolin-1 is pres ent. In other words, our prev ious work showing that AS and caveolin localized to similar crude cav eolar fractions did not prove that the localization was to the same sp ecific caveolar regions [9]. The co-localization of AS and caveolin suggests a role for caveolin-1 in assembling AS into a functional signaling microdomain, much like its role in regulati ng eNOS transport and function. For example, caveolar domains assemble at the Golgi and tr affic to the plasma membrane as stable

PAGE 198

180 transport platforms [10, 11]. These platform s constitute signaling microdomains where multiple kinases, phosphatases and regulat ory enzymes assemble into functional complexes [12, 13]. The transport mechanis ms regulating the formation of the NO metabolome at caveolae enhance NO production at the plasma membrane and involve vesicular transport elements and cy toskeletal components [10, 11, 14-17]. In addition, interaction with other proteins plays a role in eNOS targeting. For example, there are two eNOS-interacting pr oteins named NOSIP (f or eNOS-interacting protein) [18] and NOSTRIN (for eNOS-trafficking inducer) [19], which both influence the subcellular localization of eNOS. NOS IP, targets eNOS to the cytoskeleton and inactivates it [20]. NOSTRIN, when overexpressed, leads to tr anslocation of eNOS from the plasma membrane to intracellular vesicu lar structures [19], possibly involving an endocytic process [21]. Utilizing AS overexpr ession, purification and mass-spectrometry of associated proteins we did identify NOSIP as a putative binding partner of AS. This leads to the intriguing possibility that there are some shared mechanisms in AS and eNOS transport. Although these findings could not be confirmed utiliz ing IP/western, the antibodies for NOSIP that we utilized were not very good, our IP techniques needed optimization and the AS antibody available on th e market at that time did not work well for IP. With the new methods and tools we now have available, it is certainly worthwhile to pursue this mechanism. Another mechanism that regulates the sub cellular translocation of eNOS is dual acylation, an irreversible N-my ristoylation at Gly2 and reve rsible thiopalmitoylation at

PAGE 199

181 C15 and C26 [16]. These lipid functional groups are responsible for targeting eNOS to the Golgi and plasma membrane, particularly to caveolae [16, 22, 23]. Acylation deficient eNOS variants cannot translocate to the plasma membrane and this leads to alterations in eNOS activity [14]. It was intriguing to hypothesize that AS might be modified by acylation. A bioinformatic search for motifs fo r myristoylation and palmitoylation did not identify any such motifs in the AS sequence. It is still possible that AS is acylated via motifs that have yet to be described. In fact, the palmitoylatio n motif uncovered for eNOS is distinct from the motif found on ot her signaling proteins [22]. For example, the motif identified for the Src family of protei ns is MGCXXC/S while eNOS the motif for eNOS is MGXXXSC15(GL)5C26. The human AS sequence does have a glycine at position 5 and a cysteine at position 19, making is possible that lipidation does indeed occur. Although we did not examine the trans port of AS in this work, the findings we describe below do support a dynamic translocatio n of AS in endothelial involving several key interacting partners. We believe that our evidence supports a model whereby protein interactions regulate the subcellular localizat ion of AS. Caveolin-1 may ind eed be a key factor in this level of regulation. The fact that the AS protein sequen ce has a caveolin binding motif suggests that the interaction we found via im munoprecipitation may in fact be a direct interaction. In addition, previous unpublished data from our laboratory supports this hypothesis since both GST and His-tag pull down experiments demonstrated that AS can pull down caveolin-1 (Brenda Flam, unpublished results). Since this motif has been identified, it can be utilized for the design of experiments aimed directly at disrupting the

PAGE 200

182 motif and determining not only whether the inte raction is direct, but also the specific region that is involved in th e interaction. Since we have identified an AS phosphorylation site very close to that motif, it is quite possible that phosphorylation regulates this interaction. The eNOS caveolin binding motif is located in the peptide 348FPAAPFSGW-356 [24, 25]. To date, there ha ve not been any eNOS phosphorylation sites in the vicinity of the caveolin binding motif with respec t to the secondary structure. Since the entire eNOS protein ha s not been crystallized, it is difficult to know if perhaps one of the eNOS phosphorylation sites is near the caveolin bi nding motif in the 3dimensional structure. However, it is known that phosphorylation regulates the interaction between eNOS and caveolin-1, thus it is certainly possible that the mechanism involves spatial proximity of phosphorylation site s. We speculate that there is an interrelated mechanism that controls the binding of caveolin to AS and eNOS. Along with the possible regulation of AS localization and function by caveolin-1, the finding the HSP90 also co-localizes with and interacts with AS strengthens our hypothesis due to the reciprocal regulati on of NO production by these two proteins. HSP90 is involved in the activation of eNOS by allowing the interaction of eNOS with calmodulin, the phosphorylation of eNOS by Ak t and the dissociation of eNOS from caveolin-1 [26-29]. It is possible that the AS-HSP90 interaction regulates AS through similar mechanisms. Interestingly, we were able to confirm this interaction with IP/mass spec analyses.

PAGE 201

183 Several more interactions were uncov ered via IP analyses, including the interactions of AS with AL eNOS, calmodulin, PKA and PKC but the results have not been reproduced. Previous work in our lab did indicate a direct in teraction between AS, AL and eNOS (Brenda Flam, unpublished result s) and our preliminary results do support this. In addition, the interac tion of AS with PKA and PKC supports previous studies that demonstrate the presence of both of these ki nases in caveolae [30, 31]. Considering that our results show that AS is an in vitro substrate for these kinases, these interactions are certainly worth pursuing. The possible interaction of AS with calmodulin suggests regulation via calcium signaling, and our link be tween AS and VEGF supports this since VEGF stimulates NO production, in part, by increasing intracellu lar calcium [26]. Overall, the above findings strengthen our hypothesis of a dynamic and complex set of interacting proteins in the NO metabolome. Up to this point, our results were ob tained by rational investigation of AS interacting partners based on preliminary data or the known role of certain proteins in regulating NO production. Our ultimate goal, was to put together the pi eces of the nitric oxide metabolome. In order to accomplish th is, we needed a more global approach to characterize the dynamic nature of both hypothesized interactions and novel or understudied interactions. We developed a pr oteomic approach to identify all proteins that are pulled down when an AS or an eNOS antibody was used for IP analyses. We identified 85 proteins that co-immunoprecipitated with AS and/or eNOS. From those, there were 6 that stood out as ha ving a possible f unctional role.

PAGE 202

184 First, cGMP-dependent protein kinase or protein kinase G (PKG) was found to interact with both AS and eNOS. PKG is an im portant kinase in regu lating the function of nitric oxide in vascular smooth muscle. When endothelial NO is released, it acts in smooth muscle to activate soluble guanylyl cyclase (sGC), increase cGMP and activate PKG. This leads to vasodilation [32]. PKG has been also been found to have direct functions in the endothelium. For example, PKG colocalizes to caveolae in both smooth muscle and endothelium [30]. In additi on, PKG is involved in the regulation of angiogenesis in endothelial cells [33]. Cons idering the link between VEGF and AS that we established in Chapters One and Two and the well established link between VEGF and eNOS [26, 33-41], it is quite possible that this is a true functi onal interaction. This interaction was not verified by IP/western, but optimization of the IP methodology or a different methodology altogether might reveal a different result. A second interaction revealed by IP/ma ss spectrometry was dynamin. Dynamin is a large GTP-binding protein that co-localizes with and interacts with eNOS, thereby increasing its activity [42]. It targets to Go lgi membranes and also co-localizes with caveolin in caveolae [43, 44]. In addition, NOS TRIN (eNOS trafficki ng inducer) interacts with dynamin and mediates eNOS subcellular translocation [21]. In our experiments, dynamin was not found to interact with eNOS. It is possible that the conditions and/or methods used did not favor this interaction. The interaction between AS and dynamin fits well with previous, unpublished data from our lab where tandem affinity purification was utilized to pull down AS and associated prot eins. These results identified a GTP-binding protein as an AS interacting partner.

PAGE 203

185 Another AS and eNOS interacting partne r identified via IP/mass spectrometry was the Golgi SNAP receptor complex. This protein is one of several involved in transport from the endoplasmic reticulum to the Golgi and in intra-Golg i transport [45]. It has been demonstrated that in vascular disorders, there are disruptions of intracellular trafficking. Specifically, hypoxia and other c onditions cause disruptions in ER/Golgi trafficking that lead to sequestration of eNOS and a reduction in plasma membrane associated eNOS. This leads to diminished NO production [46]. The possibility of AS interacting with this protein suggests that AS trafficking in the ER and Golgi may be an important regulatory mechanism. We also identified HSP90 as an AS inte racting partner via IP/mass spec, which confirms the IP/western data described earlie r. The fact that we did not find the well documented eNOS-HSP90 interaction highlights the fact that each methodology utilized to probe for protein interactions, and the va riations within individual methodologies, can lead to false positive or false negative results. Kininogen-1 precursor was another AS and eNOS interacting partner identified via IP/mass spectrometry. This precursor is cleaved into several products including bradykinin [47]. Bradykinin is a vasodilator that regulates eNOS function via several mechanisms [48-51] and the coordinate inter action with both AS and eNOS suggests that bradykinin also plays an important role in regulating AS function.

PAGE 204

186 The final protein in the IP/mass spectro metry analyses that has an important functional link in the endothelium was prohibiti n-1. It interacted with both AS and eNOS. This protein was originally id entified in yeast and is localized to the inner mitochondrial membrane. It is now apparent th at prohibitins have diverse roles in several disease states such as obesity and inflammation [52]. Recen tly, it was determined that prohbitin-1 is highly expressed in the vascular system [5 3]. Knock-down of prohhi bitin-1 in endothelial cells leads to dysfunction characterized by increased production of reactive oxygen species in mitochondria. This suggests that prohibitin-1 is important for protecting endothelial cells from oxidative damage [53] Interestingly, prohibitin was recently found to regulate the function of OPA-1, a dyna min-like protein involved in cristae morphogenesis [54]. This regulation led to the control of cell prolif eration and apoptosis and the authors speculated a role for prohibi tin in lipid rafts. Since AS has been hypothesized to localize in mitochondria in va scular smooth muscle [1] and does localize to mitochondria in other tissues [2], it is possi ble that AS is also lo calizes in endothelial cell mitochondria. There is some evidence that eNOS localizes to mitochondria [55] but the function of the citrulline-NO cycle in endothelial mitochondria has not been studied. Furthermore, NO itself has important roles in regulating mitochondrial function and biogenesis [56, 57]. Collectively, the identi fication of dynamin and prohibitin and AS interacting partners implies important functional significance. Thus, a functional citrulline-NO cycle in mitochondria might re gulate functions that are specific to that organelle.

PAGE 205

187 Finally, another set of pr ominent interactions seen via IP, IP/mass spec and tandem affinity purification/mass spec were several cytoskeletal components including actin, vimentin and tubulin. A lthough the high expression of these proteins in most cell types leads to the initial assessment that th ese results are false positives, it is quite possible that there is an intricate cytoskeletal network responsible for the trafficking of AS within endothelial cells. Indeed, such mechanisms have been extensively characterized for eNOS. There is a prominent role of actin polymerization that regulates eNOS activity and transport [58]. More recently, it was uncovered that the protein actinin-4, an actin binding protein responsible for actin cross-linki ng, interacts with and inactivates eNOS by competitively inhibiting calcium-dependent activation [59]. Furthermore, the mechanisms of eNOS re gulation by NOSIP and NOSTRIN involve the actin cytoskeleton [18, 19]. Shear stress regu lates dilation and remodeling of resistance arteries via several cytoskeletal components such as vimentin, desmin and intermediate filaments [60]. Thus, further i nvestigation into the role of the actin cytoskeleton and other cytoskeletal components in AS-eNOS co-trans location and function is an important area for further exploration. It is clear from our studies that obtaining an accurate picture of true interacting partners is a complex process. Each technique employed can lead to false positive or false negative results. In addition, many techniques cannot confirm whether an interaction is direct or indirect. Despite th ese limitations, our studies indicat e that there is a vast and dynamic network of associating proteins that regulate the function of the nitric oxide metabolome.

PAGE 206

188 References [1] L. Xie, and S. S. Gross, Argininosucci nate synthetase overe xpression in vascular smooth muscle cells poten tiates immunostimulant-induced NO production, J Biol Chem 272 (1997) 16624-16630. [2] K. Miyanaka, T. Gotoh, A. Nagasaki, M. Takeya, M. Ozaki, K. Iwase, M. Takiguchi, K. I. Iyama, K. Tomita, and M. Mori, Immunohistochemical localizat ion of arginase II and other enzymes of arginine metabolis m in rat kidney and liver, Histochem J 30 (1998) 741-751. [3] Q. Zhang, J. E. Church, D. Jagnandan, J. D. Catravas, W. C. Sessa, and D. Fulton, Functional relevance of Golgiand plasma membrane-localized endothelial NO synthase in reconstituted endothelial cells, Arterioscler Thromb Vasc Biol 26 (2006) 1015-1021. [4] D. Fulton, R. Babbitt, S. Zoellner, J. Fontana, L. Acevedo, T. J. McCabe, Y. Iwakiri, and W. C. Sessa, Targeting of endothelial nitric -oxide synthase to the cytoplasmic face of the Golgi complex or plasma membrane re gulates Aktversus calcium-dependent mechanisms for nitric oxide release, J Biol Chem 279 (2004) 30349-30357. [5] D. Fulton, J. Fontana, G. Sowa, J. P. Gr atton, M. Lin, K. X. Li, B. Michell, B. E. Kemp, D. Rodman, and W. C. Sessa, Localizat ion of endothelial nitric-oxide synthase phosphorylated on serine 1179 and nitric oxide in Golgi and plasma membrane defines the existence of two pools of active enzyme, J Biol Chem 277 (2002) 4277-4284. [6] C. W. Shuttleworth, A. J. Burns, S. M. Ward, W. E. O'Brien, and K. M. Sanders, Recycling of L-citrulline to sustain nitric oxide-dependent enteric neurotransmission, Neuroscience 68 (1995) 1295-1304. [7] L. P. Solomonson, B. R. Fl am, L. C. Pendleton, B. L. Goodwin, and D. C. Eichler, The caveolar nitric oxide s ynthase/arginine regeneration system for NO production in endothelial cells J Exp Biol 206 (2003) 2083-2087. [8] B. L. Goodwin, L. P. So lomonson, and D. C. Eichler, Argininosuccinate synthase expression is required to maintain nitric ox ide production and cell viability in aortic endothelial cells, J Biol Chem 279 (2004) 18353-18360. [9] B. R. Flam, P. J. Hartmann, M. Harrell-B ooth, L. P. Solomonson, and D. C. Eichler, Caveolar localization of argi nine regeneration enzymes, ar gininosuccinate synthase, and lyase, with endothelial nitric oxide synthase, Nitric Oxide 5 (2001) 187-197.

PAGE 207

189 [10] F. A. Sanchez, N. B. Savalia, R. G. Dura n, B. K. Lal, M. P. Boric, and W. N. Duran, Functional significance of di fferential eNOS translocation, Am J Physiol Heart Circ Physiol 291 (2006) H1058-1064. [11] J. Liu, G. Garcia-Cardena, and W. C. Sessa, Palmitoylation of endothelial nitric oxide synthase is necessary for optimal stimulat ed release of nitric oxide: implications for caveolae localization, Biochemistry 35 (1996) 13277-13281. [12] P. W. Shaul, Regulati on of endothelial nitric oxide synthase: location, location, location, Annu Rev Physiol 64 (2002) 749-774. [13] O. Feron, and J. L. Balligand, Caveolin s and the regulation of endothelial nitric oxide synthase in the heart, Cardiovasc Res 69 (2006) 788-797. [14] E. Gonzalez, R. Kou, A. J. Lin, D. E. Golan, and T. Michel, Subcellular targeting and agonist-induced site-specifi c phosphorylation of endothelia l nitric-oxide synthase, J Biol Chem 277 (2002) 39554-39560. [15] P. Der, J. Cui, and D. K. Das, Role of lipid rafts in ceramide and nitric oxide signaling in the ischemic and preconditioned hearts, J Mol Cell Cardiol 40 (2006) 313320. [16] P. W. Shaul, E. J. Smart, L. J. Robinson, Z. German, I. S. Yuhanna, Y. Ying, R. G. Anderson, and T. Michel, Acylation targets endothelial n itric-oxide synthase to plasmalemmal caveolae, J Biol Chem 271 (1996) 6518-6522. [17] P. Prabhakar, H. S. Thatte, R. M. Go etz, M. R. Cho, D. E. Golan, and T. Michel, Receptor-regulated translocati on of endothelial nitric-oxide synthase, J Biol Chem 273 (1998) 27383-27388. [18] J. Dedio, P. Konig, P. Wohlfart, C. Sc hroeder, W. Kummer, and W. Muller-Esterl, NOSIP, a novel modulator of endothelial ni tric oxide synthase activity, Faseb J 15 (2001) 79-89. [19] K. Zimmermann, N. Opitz, J. Dedio, C. Renne, W. Muller-Esterl, and S. Oess, NOSTRIN: a protein modulating nitric oxide release and subcellu lar distribution of endothelial nitric oxide syntha se, Proc Natl Acad Sci U S A 99 (2002) 17167-17172. [20] M. Schleicher, F. Brundin, S. Gross, W. Muller-Esterl, and S. Oess, Cell cycleregulated inactivation of endot helial NO synthase through NO SIP-dependent targeting to the cytoskeleton, Mol Cell Biol 25 (2005) 8251-8258. [21] A. Icking, S. Matt, N. Opitz, A. Wies enthal, W. Muller-Esterl, and K. Schilling, NOSTRIN functions as a homotrimeric adapto r protein facilitating internalization of eNOS, J Cell Sci 118 (2005) 5059-5069.

PAGE 208

190 [22] L. J. Robinson, and T. Michel, Mutagenesis of palmitoylation sites in endothelial nitric oxide synthase identifies a novel motif for dual acylation and subcellular targeting, Proc Natl Acad Sci U S A 92 (1995) 11776-11780. [23] S. Oess, A. Icking, D. Fulton, R. Govers and W. Muller-Esterl, Subcellular targeting and trafficking of nitric oxi de synthases, Biochem J 396 (2006) 401-409. [24] H. Ju, R. Zou, V. J. Venema, and R. C. Venema, Direct intera ction of endothelial nitric-oxide synthase and caveolin-1 inhi bits synthase activity, J Biol Chem 272 (1997) 18522-18525. [25] G. Garcia-Cardena, P. Martasek, B. S. Ma sters, P. M. Skidd, J. Couet, S. Li, M. P. Lisanti, and W. C. Sessa, Dissecting the inte raction between nitric oxide synthase (NOS) and caveolin. Functional signifi cance of the nos caveolin binding domain in vivo, J Biol Chem 272 (1997) 25437-25440. [26] A. Brouet, P. Sonveaux, C. Dessy, J. L. Balligand, and O. Feron, Hsp90 ensures the transition from the early Ca2+-dependent to the late phosphorylation-dependent activation of the endothelial nitric-oxide syntha se in vascular endothe lial growth factorexposed endothelial cells, J Biol Chem 276 (2001) 32663-32669. [27] G. Garcia-Cardena, R. Fan, V. Shah, R. Sorrentino, G. Cirino, A. Papapetropoulos, and W. C. Sessa, Dynamic activation of endothelial nitric oxide synthase by Hsp90, Nature 392 (1998) 821-824. [28] J. P. Gratton, J. Fontana, D. S. O'C onnor, G. Garcia-Cardena, T. J. McCabe, and W. C. Sessa, Reconstitution of an endothelial nitric-oxide synthase (eNOS), hsp90, and caveolin-1 complex in vitro. Evidence th at hsp90 facilitates calmodulin stimulated displacement of eNOS from caveolin-1, J Biol Chem 275 (2000) 22268-22272. [29] W. Xi, H. Satoh, H. Kase, K. Suzuki and Y. Hattori, Stimulated HSP90 binding to eNOS and activation of the PI3-Akt pathway contribute to globular adiponectin-induced NO production: vasorelaxation in response to globular adiponectin, Biochem Biophys Res Commun 332 (2005) 200-205. [30] A. E. Linder, L. P. McCluskey, K. R. Cole, 3rd, K. M. Lanning, and R. C. Webb, Dynamic association of nitric oxide downs tream signaling molecules with endothelial caveolin-1 in rat aorta, J Pharmacol Exp Ther 314 (2005) 9-15. [31] C. Mineo, Y. S. Ying, C. Chapline, S. Jaken, and R. G. Anderson, Targeting of protein kinase Calpha to caveolae, J Cell Biol 141 (1998) 601-610.

PAGE 209

191 [32] T. Yamashita, S. Kawashima, Y. Ohashi, M. Ozaki, Y. Rikitake, N. Inoue, K. Hirata, H. Akita, and M. Yokoyama, Mechanisms of reduced nitric oxide/cGMP-mediated vasorelaxation in transgenic mice overexpre ssing endothelial nitr ic oxide synthase, Hypertension 36 (2000) 97-102. [33] M. Cudmore, S. Ahmad, B. Al-Ani, P. Hewett, S. Ahmed, and A. Ahmed, VEGF-E activates endothelial nitric oxide synthase to induce a ngiogenesis via cGMP and PKGindependent pathways, Bi ochem Biophys Res Commun 345 (2006) 1275-1282. [34] A. Bouloumie, V. B. Schini-Kerth, and R. Busse, Vascular endothelial growth factor up-regulates nitric oxide synthase expressi on in endothelial cells, Cardiovasc Res 41 (1999) 773-780. [35] D. Feliers, X. Chen, N. Akis, G. G. Choudhury, M. Madaio, and B. S. Kasinath, VEGF regulation of endothelial nitric oxide synthase in glomerular endothelial cells, Kidney Int 68 (2005) 1648-1659. [36] S. Jesmin, S. Zaedi, N. Shimojo, M. Iemitsu, K. Masuzawa, N. Yamaguchi, C. N. Mowa, S. Maeda, Y. Hattori, and T. Miya uchi, Endothelin antagonism normalizes VEGF signaling and cardiac function in STZ-induced diabetic rat hearts, Am J Physiol Endocrinol Metab 292 (2007) E1030-1040. [37] B. J. Michell, Z. Chen, T. Tiganis, D. Stapleton, F. Katsis, D. A. Power, A. T. Sim, and B. E. Kemp, Coordinated control of endothelial nitric-oxide synthase phosphorylation by protein kinase C and the cAMP-dependent protein kinase, J Biol Chem 276 (2001) 17625-17628. [38] A. Papapetropoulos, G. Ga rcia-Cardena, J. A. Madri, an d W. C. Sessa, Nitric oxide production contributes to the angiogenic properties of vascular endothelial growth factor in human endothelial ce lls, J Clin Invest 100 (1997) 3131-3139. [39] J. A. Reihill, M. A. Ewart, D. G. Hardie, and I. P. Salt, AMP-activated protein kinase mediates VEGF-stimulated endot helial NO production, Biochem Biophys Res Commun 354 (2007) 1084-1088. [40] T. Kobayashi, and K. Kamata, Short-term insulin treatment and aortic expressions of IGF-1 receptor and VEGF mRNA in diabetic ra ts, Am J Physiol Heart Circ Physiol 283 (2002) H1761-1768. [41] T. Tanimoto, Z. G. Jin, and B. C. Be rk, Transactivation of vascular endothelial growth factor (VEGF) receptor Flk-1/KDR is involved in sphingosine 1-phosphatestimulated phosphorylation of Akt and endotheli al nitric-oxide synthase (eNOS), J Biol Chem 277 (2002) 42997-43001.

PAGE 210

192 [42] S. Cao, J. Yao, T. J. McCabe, Q. Yao, Z. S. Katusic, W. C. Sessa, and V. Shah, Direct interaction betw een endothelial nitric-oxide synthase and dynamin-2. Implications for nitric-oxide synthase function, J Biol Chem 276 (2001) 14249-14256. [43] K. Schilling, N. Opitz, A. Wiesenthal, S. Oess, R. Tikkanen, W. Muller-Esterl, and A. Icking, Translocation of endothelial nitric -oxide synthase involves a ternary complex with caveolin-1 and NOSTRIN, Mol Biol Cell 17 (2006) 3870-3880. [44] S. Chatterjee, S. Cao, T. E. Peterson, R. D. Simari, and V. Shah, Inhibition of GTPdependent vesicle trafficking impairs internalization of plasmalemmal eNOS and cellular nitric oxide production, J Cell Sci 116 (2003) 3645-3655. [45] P. B. Sehgal, S. Mukhopadhyay, F. Xu, K. Patel, and M. Shah, Dysfunction of Golgi tethers, SNAREs, and SNAPs in monocrota line-induced pulmonary hypertension, Am J Physiol Lung Cell Mol Physiol 292 (2007) L1526-1542. [46] S. Mukhopadhyay, F. Xu, and P. B. Sehga l, Aberrant cytoplasmic sequestration of eNOS in endothelial cells after monocrota line, hypoxia, and senescence: live-cell caveolar and cytoplasmic NO imaging, Am J Physiol Heart Circ Physiol 292 (2007) H1373-1389. [47] Y. L. Guo, and R. W. Colman, Two f aces of high-molecular-weight kininogen (HK) in angiogenesis: bradykinin turns it on a nd cleaved HK (HKa) turns it off, J Thromb Haemost 3 (2005) 670-676. [48] M. B. Harris, H. Ju, V. J. Venema, H. Li ang, R. Zou, B. J. Michell, Z. P. Chen, B. E. Kemp, and R. C. Venema, Reciprocal phosphorylation and regula tion of endothelial nitric-oxide synthase in response to bradykinin stimulation, J Biol Chem 276 (2001) 16587-16591. [49] A. Parenti, L. Mo rbidelli, F. Ledda, H. J. Granger, and M. Ziche, The bradykinin/B1 receptor promotes angiogenesis by up-regulation of endogenous FGF-2 in endothelium via the nitric oxide synthase pathway, Faseb J 15 (2001) 1487-1489. [50] S. W. Bae, H. S. Kim, Y. N. Cha, Y. S. Park, S. A. J o, and I. Jo, Rapid increase in endothelial nitric oxide pr oduction by bradykinin is medi ated by protein kinase A signaling pathway, Bioche m Biophys Res Commun 306 (2003) 981-987. [51] H. Ju, V. J. Venema, M. B. Marrero, a nd R. C. Venema, Inhibitory interactions of the bradykinin B2 receptor with endothelial nitric-oxide synthase, J Biol Chem 273 (1998) 24025-24029. [52] C. Merkwirth, and T. Langer, Prohibi tin function within mitochondria: Essential roles for cell proliferation and cris tae morphogenesis, Biochim Biophys Acta (2008).

PAGE 211

193 [53] M. Schleicher, B. R. Shepherd, Y. Suar ez, C. Fernandez-Hernando, J. Yu, Y. Pan, L. M. Acevedo, G. S. Shadel, and W. C. Sessa, Prohibitin-1 maintains the angiogenic capacity of endothelial cells by regulating mitochondrial function and senescence, J Cell Biol 180 (2008) 101-112. [54] C. Merkwirth, S. Dargazanli, T. Tatsuta, S. Geimer, B. Lower, F. T. Wunderlich, J. C. von Kleist-Retzow, A. Waisman, B. West ermann, and T. Langer, Prohibitins control cell proliferation and apoptos is by regulating OPA1-depende nt cristae morphogenesis in mitochondria, Genes Dev 22 (2008) 476-488. [55] S. Gao, J. Chen, S. V. Brodsky, H. Hua ng, S. Adler, J. H. Lee, N. Dhadwal, L. Cohen-Gould, S. S. Gross, and M. S. Goligorsky, Docking of endot helial nitric oxide synthase (eNOS) to the mitochondrial outer me mbrane: a pentabasic amino acid sequence in the autoinhibitory domain of eNOS targets a proteinase K-cl eavable peptide on the cytoplasmic face of mitochondria, J Biol Chem 279 (2004) 15968-15974. [56] S. Borniquel, I. Valle, S. Cadenas, S. Lamas, and M. Monsalve, Nitric oxide regulates mitochondrial oxidative stress prot ection via the transcri ptional coactivator PGC-1alpha, Faseb J 20 (2006) 1889-1891. [57] J. D. Erusalimsky, and S. Moncada, N itric oxide and mitochondrial signaling: from physiology to pathophysiology, Arteri oscler Thromb Vasc Biol 27 (2007) 2524-2531. [58] Y. Su, D. Kondrikov, and E. R. Block, Beta-actin: a regulato r of NOS-3, Sci STKE 2007 (2007) pe52. [59] Y. Hiroi, Z. Guo, Y. Li, A. H. Be ggs, and J. K. Liao, Dynamic regulation of endothelial NOS mediated by competitive interaction with al pha-actinin-4 and calmodulin, Faseb J 22 (2008) 1450-1457. [60] L. Loufrani, and D. Henrion, Role of th e cytoskeleton in flow (shear stress)-induced dilation and remodeling in resistance arteries, Med Biol Eng Comput 46 (2008) 451-460.

PAGE 212

194 PERSPECTIVES Summary The work conducted in this dissertation ha s uncovered significant advances in our understanding of AS regulation as it relates to nitric oxide production and vascular health. In Chapter One, the regulation of AS functi on and expression was explored. The data revealed the important role of AS in the pr oduction of nitric oxide since overexpressing AS led to a significant increas e in endothelial NO production above the levels of the endogenous enzymes of the citrul line-NO cycle. We were also able to show that insulin and VEGF up-regulated AS expression a nd increased NO production while ceramide diminished AS expression. In Chapter Two, we identified and characterized the first post-translational modifications of AS in the endothelium: phosphorylation at serine, threonine and tyrosine residues. After utilizing bioinformatics to determine that AS phosphorylation was a possible AS regulati on mechanism, we focused on biological relevance. First, we identified PKA and PKC as kinases that regulate AS phosphorylation. We also showed that phosphorylation of AS by PKA and PKC is required for maximal VEGF-mediated NO produc tion. We then identified 7 different sites of phosphorylation utilizi ng a proteomics approach and demonstrated the potential biological roles of T131/S134, S180, S189 and S328. In Chapter Three, the subcellular

PAGE 213

195 localization of AS was define d in endothelial cells for the first time. In addition, our results demonstrated an overlap in localiza tion of AS with eNOS, caveolin-1 and HSP90. We were then able to characterize impor tant AS interacting partners utilizing immunoprecipitation including caveolin-1 and HSP90. We also provided the first example of the possible relationship between AS phosphorylation and protein interactions utilizing the proximity of S328 to the newl y identified caveolin-binding motif. Finally, utilizing proteomics, we were also able to id entify 85 proteins that interact with eNOS and/or AS in BAECs under basal conditions. From that list of proteins, 6 were highlighted as having known or potential roles in regulating NO production: PKG, dynamin, prohibitin, HSP90, the Golgi SNAP receptor complex and kininogen-1. Significance The work presented in this dissertation dem onstrated that AS is a central player in the regulation of NO production as evidenced by the multiple and intricate mechanisms that regulate its function. Often times, the stud ies that focus strictly on eNOS function to address the regulation of NO production only tell part of the story. Considering that most phenotypic properties of the endothelium are medi ated by NO [1], it is essential that the scientific community focus more atte ntion on looking at th e bigger picture. The key observation that in the endo thelium, AS transient overexpression enhances nitric oxide production further supports the fact that AS is the rate limiting step in the process [2-5]. Such an important di stinction supports our in itial hypothesis that

PAGE 214

196 multiple-levels of AS regulation need to be in place in order to maintain NO levels at the appropriate levels. The finding that AS overe xpression led to a loss of endothelial cell viability confirms the importance of AS as an anti-apoptotic signaling molecule [6] and implies that the specific mechanisms by whic h AS regulates this process warrant further investigation. The transcriptional regulation of AS is im portant in many of its key target tissues such as liver, immune cells and macropha ges [7-11]. Our work demonstrating the coordinate regulation of AS and eNOS expression by both positive and negative stimuli extends our understanding of th e importance of this mechanism in regulating the health of the endothelium. This also opens avenues for investigation of these observations in animal models of vascular disease. In a ddition, this work continues to add to the mounting evidence that the arginine paradox can be explained by the tightly coupled arginine regeneration system exemplifie d by the citrulline-NO cycle [12-14]. The link between AS and VEGF is the firs t demonstration of a potential role of AS in mediating angiogenesis. The possibl e direct effects of AS on the angiogenic process could have dramatic importance in understanding the globa l effects of AS on vascular health. Furthermore, this link may explain some of the possible roles for AS in cancer [15]. Additionally, our work links AS to calcium signaling thus opening up another possible regulatory avenue.

PAGE 215

197 Post-translational regulation of AS, whether by phosphorylation or protein interactions, is an essentially uncharacteri zed field of NO biology. Considering the fact that NO is such a potent mediator of w hole body metabolism [16, 17], it is certainly a rational principle that multiple mechanisms are needed to maintain the function of the system. Post-translational modifications allow for acute regulation of protein function. It is absolutely essential for AS to be able to respond quickly to the ever changing cellular need for NO, and we have shown that AS phosphorylation is one such mechanism. This type of regulation also opens up a multitude of therapeutic avenues aimed at the kinases and phosphatases that contro l the ever-changing barcode of post-translational modifications. Considering that eNOS is so intricately regulated by dynamic association with a number of other proteins [17], it makes sense that AS is part of this regulatory scheme. In addition, the continuation of our work to characterize the NO metabolome will be essential in identifying additional, nove l binding partners that have not previously been considered. The multiple possible functions of AS in the vasculature and the complex and numerous mechanisms that regulate its func tion, some presented in this dissertation and many yet to be uncovered, lead to the pros pect that many possible interventions to regulate AS function could impact a number of disease processes. For example, there is controversy related to the us efulness of arginine supplemen tation in the treatment of vascular disorders [18]. Our work suggests that this is because of the multiple functions of AS in vascular and nonvascular roles. Although ci trulline supplementation may indeed be a better option due to the unique metabolic fate of this amino acid as compared

PAGE 216

198 to arginine [19], it remains to be determin ed whether whole body supplementation will be effective since increasing the substrate for AS might have additional, unwanted effects such as enhanced NO production in tissues that do not require it or perhaps even pathogenic angiogenic consequen ces related to tumorigenesi s. Perhaps the delivery of citrulline to the specific tissues with a deficiency in NO producti on is a better option. In addition, our finding that AS is ph osphorylated by kinases that are also important for eNOS regulation improves our unde rstanding of how drugs that target some of those kinases might have a greater cardiov ascular impact than originally intended. In addition, targeting kinases that directly phos phorylate AS for therapeutic purposes might have greater efficacy and fewer side effects si nce the target is further downstream in the NO signaling cascade. Another therapeutic aven ue involves protein interactions. If a defect in the HSP90 interacti on with AS and eNOS can be co rrected therapeutically with a drugs such as insulin sensitizers that have been shown to correct this defect [20], the benefit to the vascular system could be prominent. The use of such specific therapies will depend on genomic and proteomic approaches that will allow us to understand the specific defect that is causing an indivi duals disease. Whether the therapy is pharmacological or nutritional, this type of information will allow for the implementation of individualized patient therapies that will dramatically improve our success in the prevention and treatment of the prominent a nd devastating diseases that plague global health.

PAGE 217

199 Limitations One of the main limitations of our work is the fact that it has almost entirely been carried out in tissue culture. Th ere are definite advantages to tissue culture systems since they allow for the study of single variables w ithout the influence of some compensatory mechanisms and of other tissues. These studies also allow for tissue specific mechanisms to be defined. Although tissue cultur e does not completely mimic an in vivo approach, it does allow for initial hypotheses and mechanisms to be developed that can then be used as starting points for translational work. We have begun projects to translate our work into relevant animal models with an ultimat e goal of determining the applicability of our work in humans. However, our work and that of others fits well wi th the known functions of eNOS in the vasculature which has been extensively characterized in animals and humans. Therefore, we feel strongly that our tissue culture work will have relevance for human health. Another limitation of our work relate s to the groundbreaking nature of our projects. For several of the experimental av enues that were undert aken, adequate tools were not readily available. Thus, methodologies had to be designed to make judicious use of available tools. In addition, for the continua tion of this work, new tools will have to be generated such as site-specific AS antibodi es and ultimately even tissue-specific AS knockout mouse models so that a thorough understanding of the role of AS in human diseases can be delineated. These tools, in conjunction with similar tools already

PAGE 218

200 available for eNOS, will allow for a thorough understanding of the importance of AS within the citrulline-NO cycle. Future Directions There are several areas of work that will be important in expanding the findings of this dissertation. In Chapter One, each pa thway studied requires two main areas of investigation to clearly delineate the role of AS expression in NO production. First, the role of insulin, VEGF and ceramide on AS speci fic activity needs to be determined. That will allow the direct connection between th ese biological molecules and AS function. Although this was addressed partially with the AS inhibitor studies with VEGF, enzyme assays are critical to make stronger conclusions. Second, the regulation of AS expression needs to be mechanistically defined by carry ing out experiments that will determine whether message stability or specific promoter regulation accounts for the up or downregulation of AS. It is also possible th at the mechanisms involving AS regulation by ARP (Argininosuccinate Synthase Regulat ory Protein) might be tied into the transcriptional regulation uncovered in Chapter One. The findings Chapter Two related to AS post-translational regulation by phosphorylation are indeed novel and could driv e the field of nitric oxide biology into multiple new and relevant areas of investigation. Ultimately, progress in this field will depend on definitively identifying specific sites of AS phosphorylation that are necessary for NO production and importantly, for AS specific activity. This will require continued

PAGE 219

201 optimization of proteomic approaches, the de velopment of AS enzyme assays and the generation of site-specific AS antibodies. Currently, experiments are underway to identify the specific sites phosphorylated by PKC and PKA and also to assess AS phosphorylation in the endogenous AS enzyme vi a immunoprecipitation. In addition, the possibility of other types of PTM needs to be explored such as nitrosylation, glycosylation and acylation. It has taken many, many years to get a decent understanding of these regulatory mechanisms for eNOS f unction. Even with all the progress, there are still many areas that require further research and understanding. Certai nly, it will take a comparable amount of time and concerted e ffort to get a strong understanding of how similar mechanisms regulate AS. The regulation of AS by protein interactio ns is also of great significance. A more thorough understanding will require both basi c molecular and proteomic approaches. Each individual protein interaction will have to be characterized extensively. First, the utilization of several types of methods such as pull down appr oaches will be necessary to confirm the interactions and to determine if the validated interactions are direct or indirect. Second, to further prove whether intera ctions are direct or indirect, studies will need to be designed to determine the specifi c regions of each protein involved in the interaction. We already have a good start with the caveolinAS interaction due to the identification of a caveolin-binding motif in the AS protein sequence. The biological significance of these interactions will need to be assessed by determining the role of each interaction on AS specific activity. Fina lly, the optimization and continuation of proteomic approaches to find additional memb ers of the nitric oxide metabolome will be

PAGE 220

202 essential to truly understand nitric oxide me tabolism from a global systems perspective. Furthermore, these approaches will also allow us to define the role of PTM in regulating protein interactions via the ut ilization of methods that can simultaneously identify protein interactions and PTM. We have already de signed these methodologies in conjunction with the Moffitt Proteomics Core and preliminary experiments have been initiated. Finally, the significance of our work will be strengthened by projects designed to characterize our tissue culture findings in animal models and ultimately in humans. We have such a project underway currently. We plan on beginning our translational work by utilizing streptozotocin (STZ) to induce diabetes in rats. This type 1 diabetes model will allow us to examine the impact of total insu lin deficiency on the expression of AS, AL and eNOS. This type of work will mimic our tissue culture experiments with insulin. In addition, we will have 4 treatment groups: sh am treated, STZ only, STZ plus suboptimal insulin treatment and STZ plus optimal insu lin treatment. This will allow us to study expression patterns in conditions of norm al glycemia, moderate hyperglycemia and extreme hyperglycemia. We will also be able to measure serum NO levels, the expression of other proteins that we ha ve identified as essential for AS regulation (VEGF, PKA, HSP90, etc.), and the subcellula r localization of these proteins. Therefore, we can build an in vivo model of specific dysfunction of the citrul line-NO cycle in diabetes as it relates to AS that can eventually be uti lized to carry out human studies.

PAGE 221

203 References [1] S. Moncada, and E. A. Higgs, The discovery of nitric oxide and its role in vascular biology, Br J Pharmacol 147 Suppl 1 (2006) S193-201. [2] L. Xie, and S. S. Gross, Argininosucci nate synthetase overe xpression in vascular smooth muscle cells poten tiates immunostimulant-induced NO production, J Biol Chem 272 (1997) 16624-16630. [3] L. Xie, Y. Hattori, N. Tume, and S. S. Gross, The preferred source of arginine for high-output nitric oxide synthesis in blood vessels, Semin Perinatol 24 (2000) 42-45. [4] W. C. Sessa, M. Hecker, J. A. Mitchell, and J. R. Vane, The metabolism of L-arginine and its significance for the biosynthesis of endothelium-derived relaxing factor: Lglutamine inhibits the generation of L-arginine by culture d endothelial cells, Proc Natl Acad Sci USA 87 (1990) 8607-8611. [5] A. Husson, C. Brasse-Lagnel, A. Fairand, S. Renouf, and A. Lavoinne, Argininosuccinate synthetase from the ur ea cycle to the citrulline-NO cycle, Eur J Biochem 270 (2003) 1887-1899. [6] B. L. Goodwin, L. P. So lomonson, and D. C. Eichler, Argininosuccinate synthase expression is required to maintain nitric ox ide production and cell viability in aortic endothelial cells, J Biol Chem 279 (2004) 18353-18360. [7] Y. Hattori, E. B. Campbell, and S. S. Gr oss, Argininosuccinate synthetase mRNA and activity are induced by immunostimulants in vascular smooth muscle. Role in the regeneration or arginine for nitric oxide synthesis, J Biol Chem 269 (1994) 9405-9408. [8] M. Flodstrom, A. Niemann, F. J. Bedoya S. M. Morris, Jr., and D. L. Eizirik, Expression of the citrulline-nitr ic oxide cycle in rodent and human pancreatic beta-cells: induction of argininosucci nate synthetase by cytokines, Endocrinology 136 (1995) 32003206. [9] E. Bizzoco, M. G. Vannucchi, and M. S. Faussone-Pellegrini, Transient ischemia increases neuronal nitric oxide syntha se, argininosuccinate synthetase and argininosuccinate lyase co-expression in rat striatal neurons, Exp Neurol 204 (2007) 252259. [10] T. Koga, W. Y. Zhang, T. Gotoh, S. Oyadomari, H. Tanihara, and M. Mori, Induction of citrulline-nit ric oxide (NO) cycle en zymes and NO production in immunostimulated rat RPE-J cells, Exp Eye Res 76 (2003) 15-21.

PAGE 222

204 [11] Y. Su, and E. R. Block, Hypoxia inhi bits the induction of argininosuccinate synthetase by endotoxin in lung endothelial cells, Am J Physiol 272 (1997) L934-938. [12] B. R. Flam, D. C. Eichler, and L. P. Solomonson, Endothelial nitric oxide production is tightly coupled to the citrul line-NO cycle, Nitric Oxide 17 (2007) 115-121. [13] C. W. Shuttleworth, A. J. Burns, S. M. Ward, W. E. O'Brien, and K. M. Sanders, Recycling of L-citrulline to sustain nitric oxide-dependent enteric neurotransmission, Neuroscience 68 (1995) 1295-1304. [14] L. J. Shen, K. Beloussow, and W. C. Shen, Accessibility of endothelial and inducible nitric oxide synthase to the intracellula r citrulline-arginine regeneration pathway, Biochem Pharmacol 69 (2005) 97-104. [15] B. J. Dillon, V. G. Prieto, S. A. Curl ey, C. M. Ensor, F. W. Holtsberg, J. S. Bomalaski, and M. A. Clark, Incidence and di stribution of argininosuccinate synthetase deficiency in human cancers: a method for identifying cancers sensitive to arginine deprivation, Cancer 100 (2004) 826-833. [16] E. Culotta, and D. E. Koshland, Jr., NO news is good news, Science 258 (1992) 1862-1865. [17] D. M. Dudzinski, and T. Michel, Life history of eNOS: partners and pathways, Cardiovasc Res 75 (2007) 247-260. [18] R. B. Preli, K. P. Klein, and D. M. He rrington, Vascular effects of dietary L-arginine supplementation, Athe rosclerosis 162 (2002) 1-15. [19] G. Wu, J. K. Collins, P. Perkins-Veazie, M. Siddiq, K. D. Dolan, K. A. Kelly, C. L. Heaps, and C. J. Meininger, Dietary supplementation with watermelon pomace juice enhances arginine availability and ameliorate s the metabolic syndrome in Zucker diabetic fatty rats, J Nutr 137 (2007) 2680-2685. [20] B. J. Davis, Z. Xie, B. Viollet, a nd M. H. Zou, Activation of the AMP-activated kinase by antidiabetes drug metformin stimul ates nitric oxide synthesis in vivo by promoting the association of heat shock protein 90 and endothelial n itric oxide synthase, Diabetes 55 (2006) 496-505.

PAGE 223

205 APPENDICES

PAGE 224

APPENDIX A: Related Publications 206 TROGLITAZONE UP-REGULAT ES VASCULAR ENDOTHELIAL ARGININOSUCCINATE SYNTHASE Bonnie L. Goodwin1,2, Karen D. Corbin1, Laura C. Pendleton1, Monique M. Levy1,2, Larry P. Solomonson1 and Duane C. Eichler1 Affiliations: 1Department of Molecular Medicine College of Medicine and the 2Johnnie B. Byrd, Sr. Alzheimers Center and Research Institute, University of South Florida, Tampa, FL 33612 Corresponding author: Duane C. Eichler Tel: 1 813-974-9716; Fax: 1 813 974-7357 E-mail: deichler@health.usf.edu

PAGE 225

APPENDIX A: (Continued) 207 Abstract Vascular endothelial nitric oxide (NO) production via the citrulline-NO cycle not only involves the regulation of endothelial nitric oxide s ynthase (eNOS), but also regulation of caveolar-localized endothelia l argininosuccinate s ynthase (AS), which catalyzes the rate-limiting step of the cycle. In the presen t study, we demonstrated that exposure of endothelial cells to troglitazone coordinately induced AS expression and NO production. Western blot analysis demonstrated an incr ease in AS protein expression. This increased expression was due to tran scriptional upregulati on of AS mRNA, as determined by quantitative real time RT-PCR and inhibition by 1-Dribofuranosylbenzimidazole (DRB), a transcript ional inhibitor. Reporter gene assays and EMSA analyses identified a distal PPAR response element (PPRE) ( to ) that mediated the troglitazone increase in AS expression. Overall, this study defines a novel molecular mechanism through which a th iazolidinedione (TZD ) like troglitazone supports endothelial function via the transc riptional up-regulation of AS expression.

PAGE 226

APPENDIX A: (Continued) 208 Introduction Almost all normal functions of vascular endothelial cells are dependent on or affected by the bioactivity of nitric oxid e (NO). Thus, impairment of endothelial NO production is often a common pathogenic m echanism by which cardiovascular risk factors such as hypercholesterolemia, hypert ension, smoking, homocystinemia, vascular inflammation, and diabetes mellitus promote thei r deleterious effects on the vascular wall [1]. Endothelial NO production is supported by reactions catalyzed by endothelial nitric oxide synthase (eNOS), argini nosuccinate synthase (AS) and argininosuccinate lyase (AL) which are core components of the citru lline-NO cycle [2-4]. The principal role of AS and AL catalysis is in the conversion of ci trulline to arginine, th e substrate utilized by eNOS to produce NO and citrulline. AS is rate -limiting to the citrulline-NO cycle [3, 4], and as such is required to sustain en dothelial function and viability [5]. PPAR is a member of the nuclear recep tor superfamily of ligand-activated transcription factors th at has been shown to regulate th e transcription of genes involved in lipid metabolism differentiation and cell growth [6]. Both naturally derived PPAR ligands, including a number of fatty acid metabo lites such as eicosanoid derivatives [7] and 15-deoxy12,14-prostaglandin J2 (15d-PGJ2) [8, 9], as well as synthetic ligands such as the thiazolidinediones (TZDs) have been described. The TZDs have insulin-sensitizing properties [10-12] which provide cardiovasc ular benefits [13-17] and promote flowmediated vasodilatation, in part, by stim ulating endothelial NO production via the activation of eNOS [17, 18]. Because of th ese findings, we examined whether the TZD,

PAGE 227

APPENDIX A: (Continued) 209 troglitazone, known to promote NO production [19] and vasodilatation in diabetic patients [20], may affect the efficiency of the citrulline-NO cycle via AS expression in vascular endothelial cells. Materials and Methods Cell Culture : Bovine aortic endothelial cells (BAEC) were cultured in complete Dulbeccos modified Eagles medium (1 g/ L glucose, Mediatech) containing 10% fetal bovine serum (Hyclone Laboratorie s), 100 units/ml penicillin and 100 g/ml streptomycin (Mediatech) at 37 C in an atmosphere of 5% CO2. Nitric Oxide Assay: BAEC were treated with troglitazone as indicated in DMEM (minus phenol red) plus 5% fetal bovine se rum. Aliquots (100 l) of media were removed at the indicated times and nitrite wa s measured as an indicator of cellular NO produced using a fluorometric method [21] Samples were read on a BMG Fluostar Galaxy spectrofluorometer in a 96-well plate. Data is presented as quantity of nitrite produced in pmols per mg protein. Western Blot Analysis: Following treatment with troglitazone, BAEC were harvested in 500 l PBS, centrifuged briefly and lysed in RIPA buffer. The lysate was incubated on ice for 30 minutes and protein concentration determined by BCA reagent (Pierce). Ten g of protein was electrophoresed on 415% polyacrylamide gels (Bio-Rad) and transferred onto membrane (Immobil on-P). Membranes were incubated with

PAGE 228

APPENDIX A: (Continued) 210 antibody 1:2500 anti-AS and 1:1000 anti-GAPDH (BD Transduction Labs) in 5% blocking solution in TBS-T (20 mM Tris-HCl, 137 mM NaCl, 0.1% Tween-20) and then washed in TBS-T. Membranes were subsequently incubated with horseradish peroxidase-conjugated anti-mouse antibody for 1 hour, immersed in ECL reagent (GE Healthcare) for 1 minute and then exposed to film. Band intensities were quantitated using ImageQuant software (Molecular Dynamics). RNA Isolation and Quantitative RT-PCR: Total RNA was isolated using Tri Reagent following the manufacturers instruct ions (Sigma). RNA was treated with DNase (Ambion DNA-free). Five hundred ng of RNA was reverse tran scribed using Superscript II (Invitrogen) as described previously [22]. Real time quantitative PCR was performed using AS specific primers ASL228 and ASR 278 [22]. Results were normalized to 18S rRNA. Vector Construction: Luciferase reporter constructs were designed to include the AS promoter and 5-UTR up to the AUG start codon cloned upstream of the luciferase gene. Luciferase reporter construct p3ASP 189 was described previously [23]. Left primers ASL-3075 (5-GTACCTCC ACTGAAATTGAA) and ASL-2616 (5GCACTCGAGGAAAGTCAAAGG CCATGGTG) were combined with ASRluc, (5ATAGAATGGCGCCGGGCG TTTCTTTATGTTTTTGGCGTCTTCCATCGTGACGG GTGACCAGCGGC) to amplify a deletion se ries of the AS promoter with an Xho I site on the 5 end and an Nco I site on the 3 end which were used to clone into the vector

PAGE 229

APPENDIX A: (Continued) 211 pGL3Basic (Promega) and create the v ectors p3ASP3075 and p3ASP2616, respectively. Mutations were made in the PPRE sites in p3ASP2616 using a three-way PCR method [23]. Primer PPREmut (5-GCTGGTCTTGATCTCCTGATCTCAGGTGA) was combined with primer ASRluc to amplify a fragment that contained the mutations. This PCR product was then used as a right prim er and paired with ASL-2616 to produce a second product. A third round of PCR was us ed with the second product as a template with primers ASL-2616 and ASRluc to enrich for the target. Amplified products were purified and ligated into pGL3Basic to cr eate p3ASP2616PPREmut. All constructs were verified by sequencing. Luciferase Assay Analysis : BAEC were cultured as de scribed above and plated in a 24 well plate prior to tran sfection. Experimental plasmi ds (200 ng each) and renilla control plasmid pRL-TK (50 ng) were transiently transfected into BAEC using TransitLT1 (Mirus) in serum free media. Transfected cells were cultured for 24 hours in media containing troglitazone and lysed in passive lysis buffer (Promega). Ten l lysate was assayed for luciferase and renilla activity us ing a Dual-Luciferase Reporter Assay System (Promega) according to the manufacturers instructions. Electrophoretic Mobility Shift Assay: Nuclear extracts, prepared from BAEC as described previously [23, 24], were combin ed with or without cold oligonucleotide competitors and incubated for 20 minutes at room temperature. Probes were labeled by combining equimolar amounts of co mplementary oligonucleotides (2x10-10 moles),

PAGE 230

APPENDIX A: (Continued) 212 which were heated to 70 C, and allowed to cool to room temperature slowly. The oligos were labeled using 10 l [ -32P]dCTP (3000 Ci/mmol) and Klenow enzyme. Unincorporated label was removed using Nuc Away spin columns (Ambion). The reaction mixture contained binding buffer ( 10 mM HEPES, pH 7.9, 10% glycerol, 1 mM DTT, 0.1 g/ l poly(dI:dC), 0.5 g/ l BSA and 4000 dpm/ l radiolabeled probe) and nuclear extract (5 g) in a total volume of 10 l which was incubated at 30 C for 30 minutes. Samples were loaded onto a 5% nondenaturing polyacrylamide gel and run at 180 V. Gels were dried under vacuum and exposed to film. Double-stranded oligonucleotides composed of the following sequences were used for EMSA analysis: PPRE (5-ACCTGAGGTCAGGAGTTCAAGACC-3), PPREmut (5ACCTGAGAACAGGAGAA CAAGACC-3), Sp1 site 1 (5GCTCCAGGCGGGGGCCGGGCCCGGGGGC G-3), Sp1 site 2 (5GGCCGGGCCCGGGGGCGGGGTCTGTGGCGC-3) and Sp1 site 3 (5CCGGTCACCGGCCCTGCCCCCGGGCCCTG-3). Statistical Analyses : Experimental data is expressed as the mean of experiments plus or minus the standard error of the mean. Each experiment was performed independently at least three times. Results The PPAR Ligand, Troglitazone, Increases Endothelial NO Production: To confirm that troglitazone stimulates NO production in cultured endothelial cells,

PAGE 231

APPENDIX A: (Continued) 213 confluent bovine aortic endothelial cells (BAECs) were incubated for 24 hours with increasing concentrations of this synthetic PPAR agonist. As shown in Figure 1A, a dose-dependent increase in NO production fo llowing treatment was observed up to 20 M troglitazone. The dose dependent effect was consistent with previous findings relative to the extent of NO produced [18]. Troglitazone Treatment Increases AS Expression: Since the expression of AS is necessary to support endothelial NO production [3-5, 25], we investigated whether troglitazone affected the increase in vascul ar endothelial NO producti on, at least in part, through the up-regulatio n of AS expression, or whethe r the increase in NO production was simply due to established effects on eNOS activation [17-19]. Confluent BAECs were treated with increasing concentrations of troglitazone for 24 hours and AS protein levels were determined by we stern blotting. As shown in Figure 1B & 1C, treatment with troglitazone resulted in an increase in AS protein that closel y correlated with the troglitazone dependent increase in NO production, demonstrating that this PPAR agonist does indeed support an increase in NO production through up -regulation of AS expression. To determine whether the increase in AS expression resulted from transcriptional upregulation, BAECs were grown to confluence and stimulated with troglitazone for 24 hours. RNA was prepared and quantitative re al time RT-PCR s howed that treated endothelial cells had a 3.5-fo ld increase with 20 M trogl itazone (Figure 2). This increase in AS mRNA could be inhibited by treatment with the transcriptional inhibitor

PAGE 232

APPENDIX A: (Continued) 214 1-D-ribofuranosyl-benzimidazole (DRB) suggest ing that the increase in steady-state AS mRNA levels was due to an increase in tr anscription rather than decreased AS mRNA turnover. These results also suggested that the increase in AS protein could be accounted for at the level of tr anscriptional regulation. Identification of a Putative PPRE in the Promoter of the AS Gene : In order to account for the transcriptional regulation of AS expression by troglitazone, the AS promoter was examined using luciferase repo rter gene constructs to identify regions regulated by this PPAR agonist. Previous work by others [26] and by us [27] has shown that three Sp1/3 elements in the proximal AS promoter are required for AS expression. Since PPAR agonists are known to mediate transcri ptional effects thr ough Sp1 elements [28, 29], we initially examined the involvement of the proximal promoter using a construct, p3ASP189, containing these three Sp1/ 3 elements in the first 189 bp of the AS promoter. However, transfection of the p3ASP189 construct into BAEC followed by treatment with troglitazone did not result in an increase in promoter activity (Figure 3A). Thus, the up-regulation of AS expression by troglitazone was not mediated by these Sp1 elements or other sequence elements located in the proximal promoter. Based on these findings, the search to id entify the element(s) involved in PPAR regulation was extended using a series of constructs containing increasing lengths of the AS promoter. Cells transfected with AS promoter constr ucts containing up to 2088 bp again showed no change in reporter gene act ivity in response to troglitazone treatment (data not shown). However, when a construc t containing 2616 bp of the AS promoter was

PAGE 233

APPENDIX A: (Continued) 215 transfected into BAEC, a significant increase in reporter gene expres sion was observed in response to treatment with troglitazone The construct containing 2616 bp of the promoter was activated 2.7-fold by 20 M troglitazone (Figure 3A). This comparative analysis of luciferase activity between treated and untreated transfections of the construct p3ASP2616 mapped the PPAR responsive region from -2616 to -2088 bp upstream of the transcriptional start. DNA sequence analysis identifi ed a near consensus PPAR response element (PPRE) from to bp (AGGTCAGGAGTTCA) in the p3ASP2616 construct. To verify the involvement of this element, comparative transient tr ansfection assays were performed using a construct mutated (non-func tional) in the putative PPRE and the wildtype construct. As shown in Figure 3B mutation of the putative PPRE site in p3ASP2616 completely abolished the activati ng effects of troglit azone supporting the involvement of this PPRE (-2471 to -2458 bp) in the distal region of the AS promoter. PPAR Binds to the AS PPRE: To further confirm the involvement of the PPRE in AS promoter function, we investigated whether PPAR binds to the putative AS PPRE. Electrophoretic mobility shif t assays (EMSAs) were perf ormed with oligonucleotides containing the putative sequence. Nuclear extr acts from troglitazone and untreated BAEC were mixed with [32P]-labeled AS PPRE oligonucleotides. As shown in Figure 4, troglitazone enhanced binding to the PPRE.

PAGE 234

APPENDIX A: (Continued) 216 To demonstrate specificity, excess un labeled PPREwt oligonucleotides were shown to compete, diminishing the signal of the shifted band. In contrast, addition of excess, unlabeled PPREmut oligonucleotide, with a mutation that should not allow binding and therefore shoul d not compete with [32P]-labeled AS PPRE oligonucleotides, did not diminish the specific signal. Thes e results were taken to further support the involvement of this distal PPRE in the AS promoter as the element that mediates the transcriptional upregulation by troglitazone. Discussion One mechanism by which PPAR agonists provide cardiova scular benefits is by enhancing endothelial NO production [17]. E ndothelium-derived NO is a potent chemical mediator with antiatherogenic properties, such as stimulation of vasorelaxation and repression of endothelial leukocyte adhesion mol ecules, platelet aggregation and smooth muscle cell proliferation [30-32]. Although troglitazone demonstrates vasodilator activities to lower blood pressure in diabetic patients, its precise mechanism is not well defined [20, 33, 34]. However, th ese studies suggest that tr oglitazone mediated direct effects on the vascular wall. Until now, troglitazone was thought to promote endothelial NO production through up-regulation of eNOS protein expres sion [19] or activity [18], although the mechanism was not established. This report is the first demonstration that the PPAR agonist, troglitazone, facilitates the producti on of vascular endothelial NO through the up-regulation of AS expressi on, the rate-limiting enzyme of the citrulline-NO cycle. The

PAGE 235

APPENDIX A: (Continued) 217 increase in AS protein le vels paralleled AS mRNA levels and the increased NO production. Since DRB, a transcriptional in hibitor, blocked th e induction of AS expression by troglitazone, our results indica ted that the increase in AS expression resulted from transcriptional regulation by this PPAR agonist. Therefore, we identified a distal PPRE in the AS promoter that mediated the transcriptional e ffects of troglitazone on AS expression. To our knowledge, this is th e first identification of a functional PPRE in the AS promoter. These results further support our view that the coordinate regulati on of endothelial AS expression and NO production is essential [5], and that physiologic or pharmacologic stimuli that promote or diminish endotheli al function do so not only by affecting eNOS activity or expression, but also by affecting AS expression [5]. Moreover, the results in this report contribute new and a dditional insight as to how PPAR agonists promote endothelial NO production thr ough diverse mechanisms [18, 19, 35]. For example, 15dPGJ2, a naturally occurring PPAR ligand, increases hsp90 expression which promotes eNOS activation, while ciglitazone and rosiglitazone do not, yet still increase NO production [35]. In addition, 15d-PGJ2 and rosiglitazone incr ease binding of hsp90 to eNOS to promote NO production, while cigl itazone does not. Finally, both 15d-PGJ2 and rosiglitazone, but not ciglitazone, increase phosphorylation of eNOS at ser1177, which is linked to enhanced enzyme activity [35] and increased NO production. In this report, troglitazone was found to promote NO pr oduction through the up -regulation of AS expression. This would be in addition to its reported effect on e NOS where troglitazone was shown to up-regulate eNOS expression through a mechanism independent of PPAR

PAGE 236

APPENDIX A: (Continued) 218 activation [19], or where cha nges in eNOS phosphorylation co rrelated with an increase in eNOS activity rather than expression [18]. Overall, the findings of this report dem onstrate that argininos uccinate synthase represents an additional and physiologically im portant step in the citrulline-NO cycle by which the TZD, troglitazone, promotes vascular endotheli al function. Although troglitazone was withdrawn from the market b ecause of its hepatic toxicity, the multiple mechanisms through which TZDs can improve insulin sensitivity, as well as NOdependent vasodilatation, suggests that fu rther studies with new TZD drugs may be warranted. Acknowledgements This work was supported by American Heart Association, Florida Affiliate Grant 0455228B, American Heart Association Predoctoral Fellowship Grant 0515122B, and the University of South Florida Foundation Ma ry and Walter Traskiewicz Memorial Fund.

PAGE 237

APPENDIX A: (Continued) 219 References [1] P. Vallance, and N. Chan, Endothelial func tion and nitric oxide: clinical relevance, Heart 85 (2001) 342-350. [2] A. Husson, C. Brasse-Lagnel, A. Fairand, S. Renouf, and A. Lavoinne, Argininosuccinate synthetase from the ur ea cycle to the citrulline-NO cycle, Eur J Biochem 270 (2003) 1887-1899. [3] L. Xie, Y. Hattori, N. Tume, and S. S. Gross, The preferred source of arginine for high-output nitric oxide synthesis in blood vessels, Semin Perinatol 24 (2000) 42-45. [4] L. Xie, and S. S. Gross, Argininosucci nate synthetase overe xpression in vascular smooth muscle cells poten tiates immunostimulant-induced NO production, J Biol Chem 272 (1997) 16624-16630. [5] B. L. Goodwin, L. P. So lomonson, and D. C. Eichler, Argininosuccinate synthase expression is required to maintain nitric ox ide production and cell viability in aortic endothelial cells, J Biol Chem 279 (2004) 18353-18360. [6] E. D. Rosen, and B. M. Spiegelman, PPARgamma : a nuclear regulator of metabolism, differentiation, and cell growth, J Biol Chem 276 (2001) 37731-37734. [7] S. A. Kliewer, S. S. Sundseth, S. A. Jones, P. J. Brown, G. B. Wisely, C. S. Koble, P. Devchand, W. Wahli, T. M. Willson, J. M. Lenhard, and J. M. Lehmann, Fatty acids and eicosanoids regulate gene expression thr ough direct interactions with peroxisome proliferator-activated receptors alpha and gamma, Pr oc Natl Acad Sci U S A 94 (1997) 4318-4323. [8] B. M. Forman, P. Tontonoz, J. Chen, R. P. Brun, B. M. Spiegelman, and R. M. Evans, 15-Deoxy-delta 12, 14-prostaglandin J2 is a liga nd for the adipocyte determination factor PPAR gamma, Cell 83 (1995) 803-812. [9] S. A. Kliewer, J. M. Lenhard, T. M. Willson, I. Patel, D. C. Morris, and J. M. Lehmann, A prostaglandin J2 metabolite binds pe roxisome proliferator-activated receptor gamma and promotes adipocyte differentiation, Cell 83 (1995) 813-819. [10] J. B. Majithiya, A. N. Paramar, and R. Balaraman, Pioglitazone, a PPARgamma agonist, restores endothelial f unction in aorta of streptozot ocin-induced diabetic rats, Cardiovasc Res 66 (2005) 150-161.

PAGE 238

APPENDIX A: (Continued) 220 [11] I. Kosegawa, S. Chen, T. Awata, K. Negishi, and S. Katayama, Troglitazone and metformin, but not glibenclamide, decreas e blood pressure in Otsuka Long Evans Tokushima Fatty rats, Clin Exp Hypertens 21 (1999) 199-211. [12] K. Saku, B. Zhang, T. Ohta, and K. Arakawa, Troglitazone lowers blood pressure and enhances insulin sensitivity in Watanabe heritable hyperlipidemic rabbits, Am J Hypertens 10 (1997) 1027-1033. [13] J. M. Olefsky, Treatment of insulin resistance with peroxisome proliferator-activated receptor gamma agonists, J Clin Invest 106 (2000) 467-472. [14] E. Shinohara, S. Kihara, N. Ouchi, T. Funahashi, T. Nakamura, S. Yamashita, K. Kameda-Takemura, and Y. Matsuzawa, Troglitazone suppresses intimal formation following balloon injury in in sulin-resistant Zucker fatty rats, Atherosclerosis 136 (1998) 275-279. [15] C. C. Chen, H. J. Wang, H. C. Shih, L. Y. Sheen, C. T. Chang, R. H. Chen, and T. Y. Wang, Comparison of the metabolic effects of metformin and troglitazone on fructoseinduced insulin resistance in male Sp rague-Dawley rats, J Formos Med Assoc 100 (2001) 176-180. [16] A. R. Collins, W. P. Meehan, U. Ki ntscher, S. Jackson, S. Wakino, G. Noh, W. Palinski, W. A. Hsueh, and R. E. Law, Troglitazone inhibits formation of early atherosclerotic lesions in diabetic and nondiabetic low density lipoprotein receptordeficient mice, Arterioscler Thromb Vasc Biol 21 (2001) 365-371. [17] D. S. Calnek, L. Mazzella, S. Roser, J. Roman, and C. M. Hart, Peroxisome proliferator-activated receptor gamma ligands increase release of nitric oxide from endothelial cells, Arterioscl er Thromb Vasc Biol 23 (2003) 52-57. [18] D. H. Cho, Y. J. Choi, S. A. Jo, and I. Jo, Nitric oxide production and regulation of endothelial nitric-oxide synthase phosphorylation by prolonged treatment with troglitazone: evidence for involvement of pe roxisome proliferator-activated receptor (PPAR) gamma-dependent and PPARgamma-i ndependent signaling pathways, J Biol Chem 279 (2004) 2499-2506. [19] K. Goya, S. Sumitani, M. Otsuki, X. X u, H. Yamamoto, S. Kurebayashi, H. Saito, H. Kouhara, and S. Kasayama, The thiazolidinedio ne drug troglitazone up-regulates nitric oxide synthase expression in vascular e ndothelial cells, J Diab etes Complications 20 (2006) 336-342.

PAGE 239

APPENDIX A: (Continued) 221 [20] T. Ogihara, H. Rakugi, H. Ikegami, H. Mikami, and K. Masuo, Enhancement of insulin sensitivity by troglitazone lowers blood pressure in diabetic hypertensives, Am J Hypertens 8 (1995) 316-320. [21] T. P. Misko, R. J. Schilling, D. Salvemini, W. M. Moore, and M. G. Currie, A fluorometric assay for the measurement of nitr ite in biological samples, Anal Biochem 214 (1993) 11-16. [22] L. C. Pendleton, B. L. Goodwin, B. R. Flam, L. P. Solomonson, and D. C. Eichler, Endothelial argininosuccina te synthase mRNA 5-untra nslated region diversity. Infrastructure for tissue-speci fic expression, J Biol Chem 277 (2002) 25363-25369. [23] L. C. Pendleton, B. L. Goodwin, L. P. Solomonson, and D. C. Eichler, Regulation of endothelial argininosuccinate synthase expression a nd NO production by an upstream open reading frame, J Biol Chem 280 (2005) 24252-24260. [24] C. L. Yu, D. J. Meyer, G. S. Campbell, A. C. Larner, C. Carter-Su, J. Schwartz, and R. Jove, Enhanced DNA-binding activity of a Stat3-related protein in cells transformed by the Src oncoprotein, Science 269 (1995) 81-83. [25] B. R. Flam, P. J. Hartmann, M. Harrell -Booth, L. P. Solomonson, and D. C. Eichler, Caveolar localization of argi nine regeneration enzymes, ar gininosuccinate synthase, and lyase, with endothelial nitric oxide synthase, Nitric Oxide 5 (2001) 187-197. [26] G. M. Anderson, and S. O. Freytag, Syne rgistic activation of a human promoter in vivo by transcription factor Sp1, Mol Cell Biol 11 (1991) 1935-1943. [27] B. L. Goodwin, L. C. Pendleton, M. M. Levy, L. P. Solomonson, and D. C. Eichler, Tumor necrosis factor-{alpha} reduces argini nosuccinate synthase expression and nitric oxide production in aortic endothelial cel ls, Am J Physiol Heart Circ Physiol 293 (2007) H1115-1121. [28] Y. Sassa, Y. Hata, L. P. Aiello, Y. Taniguchi, K. Kohno, and T. Ishibashi, Bifunctional properties of peroxisome prolif erator-activated receptor gamma1 in KDR gene regulation mediated via interacti on with both Sp1 and Sp3, Diabetes 53 (2004) 1222-1229. [29] A. Sugawara, A. Uruno, M. Kudo, Y. Ikeda, K. Sato, Y. Taniyama, S. Ito, and K. Takeuchi, Transcription suppression of th romboxane receptor gene by peroxisome proliferator-activated receptor-gamma via an interaction with Sp1 in vascular smooth muscle cells, J Biol Chem 277 (2002) 9676-9683. [30] U. Forstermann, E. I. Closs, J. S. Pollo ck, M. Nakane, P. Schwarz, I. Gath, and H. Kleinert, Nitric oxide synt hase isozymes. Characterizat ion, purification, molecular cloning, and functions, Hypertension 23 (1994) 1121-1131.

PAGE 240

APPENDIX A: (Continued) 222 [31] R. Joannides, W. E. Haefeli, L. Linder, V. Richard, E. H. Bakkali, C. Thuillez, and T. F. Luscher, Nitric oxide is responsi ble for flow-dependent dilatation of human peripheral conduit arteries in vivo, Circulation 91 (1995) 1314-1319. [32] S. Moncada, and A. Higgs, The L-argi nine-nitric oxide pathway, N Engl J Med 329 (1993) 2002-2012. [33] J. Kawasaki, K. Hirano, J. Nishimura, M. Fujishima, and H. Kanaide, Mechanisms of vasorelaxation induced by troglitazone, a novel antidiabetic drug, in the porcine coronary artery, Circulation 98 (1998) 2446-2452. [34] J. Song, M. F. Walsh, R. Igwe, J. L. Ram, M. Barazi, L. J. Dominguez, and J. R. Sowers, Troglitazone reduces contraction by inhibition of vascular smooth muscle cell Ca2+ currents and not en dothelial nitric oxide production, Diabetes 46 (1997) 659-664. [35] A. T. Gonon, A. Bulhak, F. Labruto, P. O. Sjoquist, and J. Pe rnow, Cardioprotection mediated by rosiglitazone, a peroxisome proliferator-activated receptor gamma ligand, in relation to nitric oxide, Basic Res Cardiol 102 (2007) 80-89.

PAGE 241

APPENDIX A: (Continued) 223 Figure Legends Figure 1: The PPAR agonist, troglitazone, stimulates endothelial NO production and AS protein expression. BAEC were treat ed with increasing concentrations of troglitazone as indicated for 24 hours. (A ) NO was measured as nitrite produced/mg protein. (Nitrite is a stable reaction product of NO and molecular oxygen.) Results are expressed as relative levels of NO produced in control (no trea tment) versus treated cells, and error bars represent the standa rd error of the mean. (B-C) Ten g of whole cell lysate was loaded onto an SDS polyacrylamide gel and standard western blotting performed. Anti-AS (1:2500) was used to detect the amount of AS protein presen t. A representative western blot is shown in B, and relative s pot density for AS protein, normalized against GAPDH and quantitated, is repres ented in C. These results are representative of three independent experiments and error bars repr esent the standard error of the mean. Figure 2: Troglitazone induces transcription of AS mRNA. BAEC were untreated (U) or treated with 20 M tr oglitazone plus or minus the transcriptional inhibitor DRB (50 M) (T and T+D, respectively) for 24 hours. Total RNA was isolated, and AS mRNA was detected using real time quantitative RT-PCR. Results were normalized to 18S rRNA and represent the average the standard error of the mean. Figure 3: Troglitazone induces a distal element in the AS promoter (A) BAEC were transiently transfected with the proximal AS promoter construct, p3ASP189, or an

PAGE 242

APPENDIX A: (Continued) 224 extended AS promoter construct, p3ASP 2616, and treated with 20 M troglitazone (Trog) for 24 hours. (B) BAEC were transiently transf ected with the AS promoter constructs with wild type p3ASP2616 (W) or p3ASP2616PPREmut (M, represents mutated PPRE) and treated with 20 M troglit azone (Trog) for 24 hours. All results are presented as relative luciferase activ ity units and represent the average the standard error of the mean of at least four experiments conducted in triplicate. Figure 4: Troglitazone increases binding to the AS PPRE. (A) Electrophoretic mobility shift assays were performed using BAEC nuc lear extracts prepared from untreated and troglitazone treated cells for 6 hours. Extracts were combined with an oligonucleotide probe containing the putative PPRE sequence of the AS promoter, and competed with either a 100-fold excess of cold wild-type or mutated oligonucleotide probe where indicated. Labeled arrow indicates position of PPAR specific bands. (B) Relative density of PPAR specific bands.

PAGE 243

APPENDIX A: (Continued) 225 Figure A-1

PAGE 244

APPENDIX A: (Continued) 226 Figure A-2

PAGE 245

APPENDIX A: (Continued) 227 Figure A-3

PAGE 246

APPENDIX A: (Continued) 228 Figure A-4

PAGE 247

ABOUT THE AUTHOR Karen Davidowitz Corbin was born in Sa n Juan, Puerto Rico. She obtained her Bachelors Degree in Nutrition and Food Scie nce from Florida State University in 1997. She completed her Dietetic Internship at the James A. Haley Veterans Hospital in 1998. From there, she spent 5 years working as a clin ical dietitian, certified diabetes educator and administrative director at The Heart and Va scular Institute of Florida as part of the LIFEHELP preventive medicine team and the CardioMAX heart failure program. Karen received her PhD in Molecular Medicine from the University of South Florida College of Medicine in 2008. Her short term goal is to conduct post-doc toral translational research related to nutrition and meta bolic disorders utilizing both basic and genomic techniques. Her long term goals include contributing to nutrition genomic research and being a catalyst for moving the professi on of dietetics securely into the genomic medicine era.