(1-7)/Mas Axis Compensates Absent Bradykinin in Bdkrb2-/- and Klkb1-/- Mice to Regulate Thrombosis Risk

ANGIOTENSIN-(1-7)/MAS AXIS COMPENSATES ABSENT BRADYKININ IN BDKRB2-/- AND KLKB1-/- MICE TO REGULATE THROMBOSIS RISK

CHAO FANG

Submitted in partial fulfillment of requirements For the degree of Doctor of Philosophy

Dissertation Advisor: Dr. Alvin Schmaier

Department of Pathology

CASE WESTERN RESERVE UNIVERSITY

January, 2014

Case Western Reserve University

School of Graduate Studies

We hereby approve the thesis/dissertation of

Chao Fang

Candidate for the Doctor of Philosophy degree*

George R. Dubyak

(Committee Chair)

Alvin H. Schmaier

Jonathan S. Stamler

Eugene Podrez

Clive D. Hamlin

Date 10/02/2013

*We also certify that written approval has been obtained for any proprietary material contained therein. Table of Contents List of Tables ...... 3 List of Figures ...... 3 Acknowledgement ...... 6 Contribution of Co-authors ...... 7 List of Abbreviations ...... 8 Abstract ...... 10 Chapter 1: Renin Angiotensin System (RAS) and thrombosis ...... 13 Plasma Kallikrein-Kinin (KKS) and Renin-Angiotensin Systems (RAS) ...... 14 KKS and contact activation ...... 14 RAS and hypertension modulation ...... 15 Interaction between KKS and RAS ...... 17 Influence of KKS, RAS and Mas receptor on thrombosis risk ...... 19 The antithrombotic nature of KKS ...... 19 The prothrombotic nature of RAS ...... 19 Mas receptor in thrombosis ...... 20 Platelet spreading and GPVI activation ...... 21 Signaling during platelet activation ...... 21 Platelet spreading ...... 24 Integrin-mediated platelet spreading ...... 25 Nitric oxide and prostacyclin as platelet inhibitors ...... 26 1. Nitric oxide: ...... 28 2. Prostacyclin: ...... 28 Overall hypothesis for the research ...... 29 Chapter 2: Angiotensin-(1-7)/Mas Decreases Thrombosis in Bdkrb2-/- Mice by Increasing NO and Prostacyclin to Reduce Platelet Spreading and GPVI Activation ...... 31 INTRODUCTION ...... 32 RESULTS ...... 34 Characterization of angiotensin-(1-7) and Mas of the renin angiotensin system in Bdkrb2-/- mice...... 34 Influence of Mas on thrombosis protection in Bdkrb2-/- mice...... 38 Platelet function of Bdkrb2-/- mice...... 41 Bdkrb2-/- platelet spreading ...... 43 GPVI activation in Bdkrb2-/- platelets ...... 47

DISCUSSION ...... 54 Chapter 3: Angiotensin-(1-7)/Mas Protects Klkb1-/- Mice from Thrombosis by Increased Prostacyclin to Reduce Vascular Tissue Factor through Elevated Sirt1 ...... 60 INTRODUCTION ...... 61 RESULTS ...... 63 Characterization of Klkb1-/- mice ...... 63 Thrombosis protection is not just defective contact activation in Klkb1-/- mice ...... 64 Role of Mas receptor in thrombosis protection in Klkb1-/- mice ...... 67 Thrombosis protection is mediated by elevated plasma PGI2 and vascular Sirt1 ...... 70 DISCUSSION ...... 76 Chapter 4: Methods and Materials ...... 82 Methods for Chapter 2 (bradykinin B2 receptor deficient mice)...... 83 Methods for Chapter 3 (prekallikrein deficient mice) ...... 93 Chapter 5: Summary and Future Directions ...... 101 Mechanisms for the compensatory effect of Angiotensin/Mas axis on Bradykinin/B2R signaling ...... 102 Mechanisms for the beneficial effect of PGI2 ...... 103 Mechanisms for reduced GPVI activation and integrin dependent spreading defect in Bdkrb2-/- mice...... 104 Molecular mechanisms by which PGI2 maintain vascular homeostasis...... 105 Conclusions ...... 105 Literature Cited ...... 106

List of Tables

Table I Primers for Real-time PCR in Bdkrb2-/- mice 86

Table II Primers for Real-time PCR in Klkb1-/- mice 98

List of Figures

Figure 1 Kallikrein Kinin System (KKS) and Contact activation 15

Figure 2 Kallikrein Kinin System (KKS), Renin Angiotensin System 18 (RAS) and their interactions

Figure 3 Platelet activation by GPVI through the immune receptor in 24 platelets

Figure 4 Nitric oxide and prostacyclin as platelet inhibitors 27

Figure 5 Influences of angiotensin receptor blocker (ARB) on the 35 Bdkrb2-/- mice

Figure 6 Angiotensin-(1-7) level in Bdkrb2-/- mice 36

Figure 7 Measurement of Ang-(1-7) formation by PRCP or ACE2 in 37 kidney from Bdkrb2+/+ and Bdkrb2-/- mice

Figure 8 Mas receptor expression in Bdkrb2-/- mice 38

Figure 9 Proposed mechanism by which elevation of angiotensin II 39 (AngII) leads to thromboprotection in Bdkrb2-/- mice

Figure 10 The influence of the receptor Mas on thrombosis risk in 40 Bdkrb2-/- mice

Figure 11 cGMP and cAMP levels in Bdkrb2-/- platelet 42

Figure 12 ADP induced fibrinogen binding in Bdkrb2-/- platelets 42

Figure 13 α- and γ- thrombin induced platelet activation in Bdkrb2-/- 43 platelets

Figure 14 Static adhesion to collagen with washed platelets 44

Figure 15 Spreading on collagen in Bdkrb2-/- platelets 45

Figure 16 Platelet spreading on GFOGER 46

Figure 17 Bdkrb2-/- platelet spreading on fibrinogen and CRP 46

Figure 18 Incubation of washed Bdkrb2-/- platelets reverses their 47 spreading defect

Figure 19 Collagen-related peptide (CRP) and convulxin (CVX) activation 47 of Bdkrb2-/- platelets

Figure 20 GPVI level in Bdkrb2-/- platelets 48

Figure 21 Influence of Mas antagonist A-779 on GPVI activation in 49 Bdkrb2-/- platelets

Figure 22 Influence of NO on GPVI activation in Bdkrb2-/- platelets 50

Figure 23 Influence of prostacyclin on GPVI activation in Bdkrb2-/- 51 platelets

Figure 24 Ligation-dependent Syk phosphorylation in Bdkrb2-/- platelets 52

Figure 25 Bone marrow transplantation experiments 53

Figure 26 Mechanism for thromboprotection in Bdkrb2-/- mice 59

Figure 27 Characterization of Klkb1-/- mice 64

Figure 28 Contact activation induced thrombin generation time (TGT) in 65 Klkb1-/-

Figure 29 Role of contact activation in thrombosis protection in Klkb1-/- 67

Figure 30 The influence of prekallikrein deficiency on the RAS and KKS 67

Figure 31 The influence of Mas receptor on thrombosis risk in Klkb1-/- 69 mice

Figure 32 Renal ACE2 and PRCP activity in Klkb1+/+ and Klkb1-/- mice 70

Figure 33 Plasma prostacyclin levels in Klkb1+/+ and Klkb1-/- mice 70

Figure 34 Platelet function in Klkb1+/+ and Klkb1-/- mice 71

Figure 35 mRNA expression of vascular-specific genes in aortic tissues 73 from Klkb1+/+ and Klkb1-/- mice

Figure 36 Endogenous thrombin potential induced tissue factor in Klkb1-/- 74 mice

Figure 37 Mechanisms of thrombosis protection in Klkb1-/- mice 77

Acknowledgement

Without the support from my parents, colleagues and friends, this work would not have been completed. I would like to thank to my advisor Dr. Alvin Schmaier, who has been supportive and patient throughout the period of my Phd studies.

Your knowledge and experience helped me grow up towards a scientist with independent thinking. Your patience encouraged me to fearlessly face any problems. Our daily discussion has influenced my view of science and constantly sparks new ideas. I had a great PhD experience. I also would like to thank to all the co-workers in the lab. It is my great pleasure to work with all of you over the years. Special thanks to Evi Stavrou, who has been helping and encouraging me on many aspects. I will definitely miss the bagels you bring to the lab meeting on every Tuesday morning. I am also grateful to all of the collaborators, who provided technical support for my research. I also want to thank my PhD committee members, Drs. George Dubyak, Jonathan Stamler, Eugene Podrez, and Clive Hamlin, whose constructive criticism has helped my projects moving forward. And last, I would like to express my deepest gratitude to my family, my parents and Jinyu Wang, who became my wife on my birthday 2013. You have been giving me hope and love throughout my Phd studies.

Contribution of Co-authors

The following co-authors provided significant amount of technical support:

Evi X. Stavrou (Chapter 2 & 3)

Nadja Grobe (Chapter 2 & 3)

Alec A. Schmaier (Chapter 2)

Andrew Chen (Chapter 2)

Marvin T. Nieman (Chapter 2)

Gregory N. Adams (Chapter 2)

Gretchen LaRusch (Chapter 2)

Yihua Zhou (Chapter 2)

Matthew L. Bilodeau (Chapter 2)

Yunmei Wang (Chapter 3)

List of Abbreviations

ACE: angiotensin converting enzyme

ACE2: angiotensin converting enzyme-2

A.U.: arbitrary units

Ang II: angiotensin II

Ang-(1-7): Angiotensin (1-7)

APTT: activated partial thromboplastin time

ASO: antisense oligonucleotide

AT1R: angiotensin type 1 receptor

AT2R: angiotensin type 2 receptor

AUC: area under curve

B1R: bradykinin B1 receptor

B2R: bradykinin B2 receptor cAMP cyclic adenosine monophosphate cGMP: cyclic guanosine monophosphate cPGI2: carbaprostacyclin

CRP: collagen related peptides eNOS: endothelial nitric oxide synthase

FXI: factor XI (zymogen)

FXIa: activated factor XI (enzyme)

FXII: factor XII (zymogen)

FXIIa: activated factor XII (enzyme)

GAPDH: glyceraldehyde 3-phosphate dehydrogenase

GPCR: G-protein coupled receptor

GPVI: glycoprotein VI

HK: high molecular weight kininogen

HK/PK: high molecular weight kininogen/plasma prekallikrein complex

IBMX: 1-methyl-3-(2-methylpropyl)-7H-purine-2,6-dione

KKS: kallikrein-kinin system

KLF2: kruppel-like factor 2

KLF4: kruppel-like factor 4

NO: nitric oxide

PAI-1: plasminogen activator inhibitor-1

PARs: protease-activated receptors 1

PGI2: prostacyclin

PK: prekallikrein

PRCP: prolylcarboxypeptidase

PT: prothrombin time

RAS: renin-angiotensin system

SEM: standard error of the mean

TF: tissue factor

TGT: thrombin generation time

TM: thrombomodulin tPA: tissue type plasminogen activator

Angiotensin-(1-7)/Mas Axis Compensates Absent Bradykinin in Bdkrb2-/- and Klkb1-/- Mice to Regulate Thrombosis Risk

Abstract

CHAO FANG

Part I:

Bradykinin B2 receptor deleted mice (Bdkrb2-/-) have delayed carotid artery thrombosis times and prolonged tail bleeding times due to elevated angiotensin II

(AngII) and angiotensin receptor 2 (AT2R) producing increased plasma nitric oxide and prostacyclin (Blood 2006;108:192). Bdkrb2-/- also have elevated plasma angiotensin-(1-7) and mRNA and protein for its receptor Mas. Blockade of Mas with its antagonist A-779 shortens Bdkrb2-/- thrombosis times, bleeding

-/- times and lowers plasma nitrate, and 6-keto-PGF1α. Bdkrb2 platelets express increased NO, cGMP and cAMP. They have reduced spreading on collagen and peptide GFOGER that is corrected by in vivo A-779 or combined L-NAME and nimesulide treatment. Bdkrb2-/- platelets have reduced CRP-induced integrin

α2bβ3 activation and P-selectin expression that are corrected by A-779 or nimesulide treatment. Transplantation of wild type bone marrow into Bdkrb2-/- hosts produces platelets with a spreading defect and delayed thrombosis times.

Transplantation of Bdkrb2-/- bone marrow into wild type hosts produces platelets with normal spreading and thrombosis times. In Bdkrb2-/- combined AT2R and

Mas over-expression produce thrombosis delay due to elevated prostacyclin and

NO and reduced collagen-induced platelet activation. In vivo, vascular Mas and

AT2R levels increase upon bradykinin B2 receptor deletion to reduce arterial thrombosis risk.

Part II:

Prekallikrein (PK) is the precursor for plasma kallikrein (KK). Recent studies show that PK deficiency is associated with reduced thrombosis risk without affecting hemostasis. However, the precise in vivo mechanism for thrombosis protection in Klkb1-/- mice is not known. Our investigations show that Klkb1-/- mice have delayed carotid occlusion time on the rose bengal and ferric chloride thrombosis model. Klkb1-/- plasma has defective contact activation-induced thrombin generation time (TGT) that partially corrects upon prolonged incubation.

However, in two models of lethal contact activation-induced pulmonary thromboembolism by collagen/epinephrine or long chain polyphosphate, Klkb1-/- mice, unlike F12-/- mice, do not have a survival advantage over WT mice,. Klkb1-/- mice have reduced plasma BK levels and reduced renal B2R mRNA. Similar to our observation in Bdkrb2-/- mice, Klkb1-/- mice have a compensatory overexpression of renal Mas receptor. Mas antagonist A-779 shortens thrombosis times in Klkb1-/- to normal. The thromboprotective mechanism in Klkb1-/- is mediated by increased plasma prostacyclin producing elevated aortic sirtuin-1

(Sirt1) and KLF-4 with down-regulated aortic tissue factor (TF). Treatment of mouse cardiac endothelial cells (MCECs) with carbaprostacyclin increased mRNA and antigen of Sirt1 and KLF4. In addition, there was decreased

11 endogenous TF-induced TGT in the Klkb1-/- plasma in the presence of infestin-4, a XIIa inhibitor. In summary, our investigation reveals a novel mechanism for thrombosis protection in Klkb1-/- mice independent of contact activation that is mediated by the receptor Mas and its product prostacyclin through elevation of vasoprotective transcription factors.

The research in these two animal models demonstrates that Ang-(1-7)/Mas axis is an important regulator of thrombosis risk. Alterations in the KKS influence thrombosis risk through receptors of the RAS.

Chapter 1: Renin Angiotensin System (RAS) and thrombosis

Plasma Kallikrein-Kinin (KKS) and Renin-Angiotensin Systems (RAS)

KKS and contact activation

The plasma kallikrein-kinin system (KKS) has a broad spectrum of physiological and pathological functions. It was originally discovered to initiate “contact activation” which leads to thrombin generation through three elements: high molecular weight kininogen (HK), prekallikrein (PK) and Factor XII (FXII) 1.

Zymogen FXII auto-activates on negatively charged surfaces such as polyphosphate (PolyP) released from bacteria or cells to generate active FXII

(FXIIa) 2. FXIIa activates PK producing plasma kallikrein which is a potent FXII activator 3. In circulation HK-bound PK is also activated on endothelial cells independent of FXIIa by prolylcarboxypeptidase (PRCP) 4. This reciprocal activation of FXII and PK results in the subsequent activation of factor XI (FXI) by

FXIIa in the intrinsic pathway of the coagulation cascade leading to thrombin generation (Figure1).

Besides its role in the coagulation cascade, FXII activation also leads to bradykinin (BK) release from HK through cleavage by kallikrein in a two-step process 5,6 (Figure1). BK is the major effector in the KKS. This nonapeptide has a sequence Arg-Pro-Pro-Gly-Phe-Ser-Pro-Phe-Arg and it targets two G-protein coupled receptors (GPCRs) on cell surfaces: B1 receptor (B1R) and B2 receptor

(B2R). B2R is constitutively expressed on a variety of cell types, while B1R is only expressed during inflammation and tissue injury 7. Upon binding to these receptors, BK has been recognized as a potent pro-inflammatory hormone with a wide range of physiological and pathological effects including vasodilation to

14 lower blood pressure, increasing vascular permeability, promoting nitric oxide

(NO) and prostacyclin synthesis, and inducing the release of tissue-type plasminogen activator (t-PA) 8-10. BK is degraded by angiotensin converting enzyme (ACE) into a stable metabolite bradykinin 1-5, which has a sequence

Arg-Pro-Pro-Gly-Phe 11. Both bradykinin and bradykinin 1-5 have been described to inhibit -thrombin-induced platelet activation by directly binding to -thrombin’s active site12. Based on these features of BK, it is recognized as a cardioprotective agent with vasodilatory and anticoagulant effects which maintain the patency of the vessel wall.

RAS and hypertension modulation

The Renin Angiotensin System (RAS) is another hormone system regulating vascular tone and counter-balances with the Kallikrein-Kinin system. Angiotensin

II (AngII) in the RAS is a major peptide that regulates blood pressure. During blood pressure regulation, the protease renin is released from the kidney into the

15 circulation to cleave the precursor protein angiotensinogen synthesized in the liver to generate a decapeptide angiotensin I (AngI). AngI is then rapidly converted to an octapeptide angiotensin II (AngII) by ACE enzyme 13. AngII binds to two GPCRs on the cell surface: angiotensin type 1 receptor (AT1R) and type 2 receptor (AT2R). AT1R activation leads to vasoconstriction in vascular smooth muscle cells (VSMC) to increase blood pressure and increase expression of procoagulant proteins plasminogen activator inhibitor-1 (PAI-1) and tissue factor

(TF) in endothelial cells 14-16. AT2R has been shown to antagonize the activity of

AT1R. Activation of AT2R results in the production of nitric oxide and the lowering of blood pressure through the NO/cGMP dependent pathway 17,18

(Figure 2). Under physiological conditions, the effect of AngII is mainly mediated through AT1R because AT2R is expressed at relatively low levels in adult tissues.

However, overexpressed AT2R is implicated in some pathological states, such as heart failure, cardiac fibrosis, stroke, renal diseases, type2 diabetes, and atherosclerosis 18. Based on these combined characteristics, AngII is considered a vasoconstrictive hormone that induces hypertension and associated vascular events such as myocardial infarction and stroke 19. A large class of antihypertensive drug called ARBs (angiotensin receptor blockers), that function to reduce the activating effects of AngII on AT1R, have been shown to reduce the risk of cardiovascular diseases 20. The breakdown product of AngII by angiotensin converting enzyme 2 (ACE2) and PRCP is angiotensin-(1-7) (Ang-(1-

7)), that binds to the GPCR Mas receptor to stimulate NO and prostacyclin production and endothelial-dependent vasodilation 21-24 (Figure 2). The Ang-(1-

7)/Mas axis along with the AngII/AT2R counter-regulates the physiological effects of AngII/AT1R signaling on vascular cells 25.

Interaction between KKS and RAS

KKS and RAS interact with each other at multiple layers. First, AngII counterbalances the influence of BK at physiological levels 26. As described above, AngII/AT1R activation results in vasoconstriction to elevate blood pressure, whereas stimulation of BK/B2R and AngII/AT2R leads to NO and prostacyclin mediated vasodilation to lower blood pressure. Some studies indicate that Ang-(1-7) and BK have synergistic effects in the blood vessels 27.

The hypotensive effect of BK is potentiated by Ang-(1-7) whereas blocking B2R with HOE 140 partially attenuates the activity of Ang-(1-7) 28,29. Second, several enzymes serve as junction points that participate in both RAS and KKS. For example, ACE not only converts AngI into AngII but also metabolizes BK into inactive BK-(1-5) (Figure 2). The clinical efficacy of anti-hypertensive medication with ACE inhibitors (ACEI) has been attributed to increased BK and reduced

AngII level by inhibited ACE activity 30,31. PRCP is another example. It was originally characterized to degrade AngII into Ang-(1-7) along with ACE2, but it was recently identified to activate PK on endothelial cells independent of FXIIa 4

(Figure 2). The observation has led to the speculation that PRCP contributes to blood pressure lowering by elevating NO and prostacyclin through increased

Ang-(1-7) and BK production. However, recent investigations using PRCP “gene trapped” mice (PRCPgt/gt) showed that decreased PRCP expression induces hypertension and increased thrombosis risk in vivo by increased reactive oxygen

17 species (ROS) in the vessels 32. This observation indicates an internal interaction between the hypertension modifying system and thrombosis risk. Third, G-protein coupled receptors (GPCRs) from the two systems interact with each other by forming homodimers and heterodimers. The functional heterodimers between

AT2R and B2R enhances the NO production from these two receptors. AT1R and B2R constitutively form heterodimers that enhance the stimulatory effect of

AngII on AT1R but does not significantly affect the signaling of BK/B2R 33-36.

Influence of KKS, RAS and Mas receptor on thrombosis risk

The antithrombotic nature of KKS

Although the KKS and RAS have been well characterized to regulate blood pressure, accumulating evidence from in vitro and in vivo studies suggest that they influence thrombosis risk. Despite its role to initiate contact activation, the

KKS is considered to be antithrombotic based on the ability of BK to induce NO and prostacyclin production and t-PA release 8,9. In addition, KKS is also involved in fibrinolysis via HK bound PK, whose activation leads to single-chain urokinase formation on endothelial cells and subsequent plasmin formation, which dissolves the fibrin components of the coagulation 37. Recently, several in vivo studies using gene depleted animals also suggest that the KKS influences thrombosis risk. Mice lacking kininogen (Kng) and HK are protected from induced arterial thrombosis and ischemic neurodegradation with reduced brain infarction in the stroke model 38,39. Our lab previously showed that depletion of B2R in mice in the B6/129 background leads to thromboprotection through a paradoxical mechanism involving receptors from the RAS 40. This study suggests that the

KKS influences thrombosis risk through the RAS independent of contact activation.

The prothrombotic nature of RAS

The prothrombotic nature of the RAS is evidenced by major clinical trials which show that treatment with ACE inhibitors (ACEI) and angiotensin receptor blockers (ARB) is associated with 15% reduction of cardiovascular risks such as myocardial infarction and stroke 41,42. In vitro evidence includes the observation

19 that AngII induces mRNA and protein expression of procoagulant TF and PAI-1 in tissue cultures of endothelial cells and astrocytes 15,16,43. Several in vivo studies showed that AngII infusion accelerates microvascular and arterial thrombosis and reduces fibrinolytic potential associated with increased plasma

PAI-1 level 44-46. However, there are no in vivo studies with gene deleted animals showing the precise influences of RAS on thrombosis risk.

Mas receptor in thrombosis

Recently accumulating evidence points to the Ang-(1-7)/Mas axis as a potential therapeutic target for thrombosis 47. As described above, Ang-(1-7) through endothelial Mas receptor stimulates prostacyclin formation and NO release in vasculature via a PI3Kinase/Akt/eNOS dependent pathway 23,24. Both NO and prostacyclin are potent inhibitors of platelet aggregation 48. The initial study demonstrating the antithrombotic effect of Ang-(1-7) was performed by

Kucharewicz et al, who showed infusion of Ang-(1-7) reduced thrombus weight in kidney-clip hypertensive rats 49. Incubation of platelets with high doses of Ang-(1-

7) (1-10 M) slightly decreased platelet adhesion to fibrillar collagen. Rajendran et al performed ex vivo experiments showing that Ang-(1-7) alone did not have any effect on platelet aggregation induced by the thromboxane A2 analogue

U46619, but potentiated the anti-aggregatory effect of the NO donor sodium nitroprusside (SNP) 50. Kucharewicz’s following study demonstrated the antithrombotic effect of Ang-(1-7) is mediated by the Mas receptor through NO and prostacyclin. They showed that Mas receptor antagonist A-779, or concomitant administration of NO synthase and prostacyclin synthase inhibitors,

20 completely abolished the effect of Ang-(1-7) on thrombosis risk in hypertensive rats 51. Fraga-Silva et al further determined the role of Mas in thrombosis using

Mas depleted mice 52. They found Ang-(1-7) specifically binds to platelet Mas receptor to stimulate NO production in platelets. This process was abolished by

Mas depletion. More importantly, Mas deficient mice have significantly shorter tail bleeding times, suggesting an important role of Mas in hemostasis 52. These combined studies indicate the antithrombotic effect of the Ang-(1-7)/Mas axis may involve prostacyclin and NO derived from both the vasculature and platelets.

In accordance with this line of thought, Fraga-Silva et al further showed that

ACE2 activation, by a small molecule XNT, produced prolonged occlusion time in spontaneously hypertensive rats (SHR) 21,53. In the follow up study, the authors developed an orally active formulation (Ang-(1-7)-CyD) by embedding Ang-(1-7) into a digestion resistant vehicle cyclodextrin 54. Oral administration of Ang-(1-7)-

CyD significantly decreased the thrombus weight in mice in a Mas receptor dependent manner, suggesting a therapeutic potential for the Ang-(1-7)/Mas axis in thrombotic diseases. Nevertheless, based on current understanding the precise mechanisms of how Ang-(1-7)/Mas influences platelet function through

NO and prostacyclin, and the relative contribution of platelet Mas and vascular

Mas is unclear and needs further investigation.

Platelet spreading and GPVI activation

Signaling during platelet activation

Platelets are critical for normal hemostasis but also contribute to important physiological and pathologic conditions such as lymphatic development, arterial

21 thrombosis, metastasis, vascular inflammation and atherosclerosis 55-59. Multiple pathways activate platelets, including: soluble agonists such as thrombin, adenosine diphosphate (ADP), epinephrine, and thromboxane A2. These agonists stimulate platelets through specific G-protein coupled receptors

(GPCRs), whereas platelet activation and spreading on adhesive molecules such as collagen, fibrinogen and von Willabrand factor (vWF) is mediated by integrins and glycoproteins 60. Platelet activation is a synergistic and complex process involving several key steps. During normal hemostasis, flowing platelets are initially captured on the subendothelial matrix by the interaction of the glycoprotein (GP) Ib-V-IX complex with von Willebrand factor (vWF) immobilized on exposed collagen. Platelet GPVI and 2 1 integrins then bind to their ligand collagen and platelet activation is initiated and then further reinforced by locally produced thrombin and soluble mediators such as ADP and serotonin released from platelets. As adhesion molecules the 1 and 3 integrins on activated platelets then shift to a high affinity state, thereby enabling them to bind their ligands and to mediate firm adhesion and spreading 61.

The signaling events initiated by different agonists converge into some common signaling pathways. Thrombin activates platelets via protease-activated receptors

(PAR-1 and PAR-4 for human, PAR-3 and PAR-4 for mouse) through Gq coupled signaling, which leads to phospholipase C (PLC ) activation 62.

Activated PLC catalyzes the hydrolysis of phosphoinositide-4,5-bisphosphate

(PIP2) to release inositol-1,4,5-trisphosphate (IP3) and 1,2-diacyl-glycerol (DAG), which together elevate cytosolic calcium levels in platelets 63. In addition,

22 integrated signals from calcium and DAG leads to activation of a small GTPase,

Rap1, which is important for integrin function 64. ADP stimulates platelets through

Gi coupled P2Y12 and Gq coupled P2Y1 receptors 65. ADP-induced fibrinogen binding depends on integrin “inside-out” activation mediated by Rap1, whose activation involves both Gi coupled phosphatidylinositol 3-kinase (PI3K) signaling and Gq coupled calcium signaling 66,67.

Collagen-induced platelet activation involves 2 receptors: GPVI and integrin

68 2b 1. They activate platelets in a reciprocal two-receptor model . Different from

GPCRs for thrombin or ADP, GPVI signals through the immune receptor pathway

(Figure 3). Upon clustering of GPVI by its ligands such as collagen or convulxin,

Src family kinases (SFKs), such as Lyn and Fyn, phosphorylate immunotyrosine activating motifs (ITAM) on the GPVI associated FcR chain 69,70. Activated ITAM induces a Syk-dependent signaling cascade leading to the activation of PLC 2, which is a common target for both GPVI and GPCR induced platelet activation pathways 69. PLC 2 activation results in calcium elevation, integrin activation, and granule secretion 71,72. Similar to Gq dependent signaling, GPVI stimulation also promotes the integrin 2b 1 inside-out activation through a small GTPase

73 Rap1b . The primed 2b 1 integrin then binds to collagen to enhance the GPVI pathway by stimulating many of the proteins in the GPVI-FcR chain cascade.

Combined signals from GPVI and 2b 1 integrin further elevate calcium level and activate rac1 to promote platelet spreading 71,74,75.

Platelet spreading

A major outcome of platelet activation is spreading, which is characterized by the protrusion of filopodia and lamellipodia, physiological structures that control the first key steps in hemostasis by mediating platelet adhesiveness and aggregation. By extending filopodia and forming lamellipodia, spreading platelets increase their surface area and thus strengthen the contact with the matrix surface and other platelets. This process is important for the following aggregation and the establishment of a stable plug 76. Thus, platelet spreading plays an important role in hemostasis.

The polymerization of actin is at the origin of platelet spreading and is responsible for the formation of protruding filopodia and lamellipodia. The assembly of actin occurs via a combined process of uncapping, severing, and

24 nucleation by adding monomers onto the growing filament (F-actin) at the barbed ends 77. This reaction is highly controlled and involves a number of actin-related regulatory proteins. Although the signaling pathways for platelet activation vary upon different stimuli, elevated calcium plays the central role in platelet spreading. Rho family small G proteins (rac1, cdc42 and RhoA) have been shown to transduce platelet activation signaling into actin assembly 76-80. Active cdc42/rac1 promotes platelet spreading while the other Rho family protein RhoA is important for platelet retraction 81. Src family kinase cSrc activation inhibits

RhoA but does not affect cdc42 and rac activity 82.

Integrin-mediated platelet spreading

Integrin signaling is bidirectional during platelet spreading: it shifts to a state with higher binding affinity for the spreading matrix (inside-out activation), which enables the following outside-in activation associated with platelet shape change.

Integrin 2b 3-mediated spreading on fibrinogen is through a similar pathway as

71,83,84 2 1-mediated spreading on collagen or its specific peptide GFOGER .

Integrin inside-out activation is promoted by Kindlin3 and Talin, which are recruited by Rap1b upon agonist stimulation as described above 64,73,85-87.

Primed integrin 2b 3 binds to its ligand fibrinogen and initiates the outside-in signaling. cSrc kinase is activated by direct interaction with activated 3 integrin at the phosphorylated cytoplasmic domain 88. Active cSrc inhibits RhoA activity, thus facilitating platelet spreading 83. In spreading platelets, the calcium dependent protease calpain gets activated following calcium elevation, and the dephosphorylated integrin cytoplasmic domain becomes susceptible to calpain

25 cleavage 89. Cleavage by calpain at the cSrc binding site removes cSrc from the

β3 tail. Deactivated cSrc releases the local inhibitory effect on RhoA activity, thereby switching the integrin signal from mediating platelet spreading to

83 retraction . More recently, the G protein subunit G 13 was demonstrated to directly interact with β3 tail to promote cSrc phosphorylation, and thus facilitate platelet spreading 90.

Nitric oxide and prostacyclin as platelet inhibitors

Nitric oxide and prostacyclin are potent vasodilators but also platelet inhibitors.

The major enzyme generating NO from its substrate L-arginine is nitric oxide synthase (NOS), which includes three isoforms: endothelial NOS (eNOS), inducible NOS (iNOS) and neuronal NOS (nNOS) 91. eNOS is constitutively expressed in platelets and endothelial cells 92. NO is a small molecule that, once generated, quickly diffuses through the plasma membrane of endothelial cells into the blood stream, where it enters the platelet to directly activate soluble guanylyl cyclase (sGC) to produce cyclic guanosine monophosphate (cGMP)

93,94 . Prostacyclin (prostaglandin I2, PGI2) is a member of the prostanoids family.

It is generated by prostacyclin synthase (PGIS) from prostaglandin H2 (PGH2), which is produced by the enzyme cyclooxygenase (COX) from a common substrate arachidonic acid 95. Unlike NO, prostacyclin has a G-protein coupled receptor (IP) on the platelet surface. Once released into the blood stream from vasculature, prostacyclin binds to the platelet IP receptor to activate adenylyl cyclase (AC) via the Gs mediated pathway to elevate intraplatelet cAMP levels

96,97 (Figure 4).

Elevated cGMP and cAMP levels in platelets activate cGMP and cAMP dependent protein kinases (PKG and PKA, respectively) 98. PKG and PKA have a broad spectrum of substrates, such as Rap1b, IP3 receptor (IP3-R), regulator of

G-protein signaling18 (RGS18) and vasodilator-stimulated phosphoprotein

(VASP) etc., whose phosphorylation results in synergistic inhibition of platelet activation including defective integrin activation, reduced calcium mobilization and inhibition of cytoskeletal actin rearrangement 98-102. In addition, recent studies suggest that NO also inhibits platelet activation through a cGMP independent mechanism by protein S-nitrosylation 103-105 (Figure 4).

A number of studies using pharmacological approaches have shown the inhibitory effect of NO and prostacyclin on platelet activation 106-111. Not until

27 recently, however, have studies with gene-depleted animals began to reveal in vivo roles of NO and prostacyclin in thrombosis and hemostasis:

1. Nitric oxide: Freedman et al showed that eNOS depleted mice have significantly shortened bleeding due to a lack of platelet-derived NO, which endogenously inhibits platelet recruitment during hemostatic responses 112. A study from a different group, however, suggests that eNOS deficiency only minimally affects hemostasis in mice with slightly reduced bleeding and thrombosis times 113. A more recent study by Zhang et al using conditional knockout mice with sGC specifically depleted in platelets raised a different hypothesis regarding the role of NO: physiological doses of NO synthesized within the platelets stimulates platelet activation, whereas higher doses of NO produced exogenously from the vasculature or NO donor inhibits platelet activation 105. Despite the controversy over the role of platelet-derived NO, it is clear that exogenous NO inhibits platelet activation. Conditional knockout animals with eNOS specifically depleted in the vasculature will be useful to further address this question 114.

2. Prostacyclin: Cheng et al reported that depletion of prostacyclin IP receptor in mice resulted in enhanced platelet activation This effect is reversed by thromboxane A2 (TxA2) receptor TP depletion, suggesting the inhibitory effect of

115 prostacyclin on platelets involves TxA2 inhibition . The following study from the same group showed that specific depletion of COX-2 in the vasculature leads to attenuated prostacyclin formation and disrupted eNOS expression and reduced

NO production, which are associated with increased risk of hypertension and

28 thrombosis in the whole animals 116. Seta et al generated a COX-2 knockdown mice which have hypomorphic COX-2 expression leading to enhanced Ferric

Chloride induced thrombosis in the cremaster arterioles 117. The antithrombotic property of PGI2 also involves a vascular effect: Riehl et al reported that COX-

1(+/-)COX-2(-/-) mice have accelerated carotid artery thrombosis due to increased procoagulant PAI-1 expression in the plasma and vasculature 118.

Barbieri et al showed that COX-2 reduces arterial thrombosis risk by inhibiting TF expression in the vasculature through elevated expression of Sirtuin-1 (Sirt1)

119,120.

Overall hypothesis for the research

In summary, it is promising for future studies to focus on the interaction between hypertension modifying systems and thrombosis. Such investigation has the potential to reveal novel mechanisms by which the KKS system influences thrombosis risk independent of contact activation by regulating platelet and/or vascular functions through the eNOS/NO and COX-2/PGI2 dependent pathways.

We hypothesized that Angiotensin-(1-7)/Mas axis can compensate absent

Bradykinin/B2R signaling in vivo. This compensation produces elevated PGI2 in the plasma to reduce thrombosis risk.

The following Chapters describe the mechanisms of thrombosis protection in two animal models with altered components in the KKS (Bradykinin B2 receptor depleted, Bdkrb2-/- in Chapter 2; prekallikrein depleted, Klkb1-/- in Chapter 3), both of which have defective bradykinin/B2R signaling leading to compensatory

29 enhanced Ang-(1-7)/Mas activity in the RAS producing increased NO and PGI2.

Using these two animal models, we have identified a mechanism that balances the KKS and RAS through bradykinin/B2R axis and Ang-(1-7)/Mas axis to regulate thrombosis risk.

Chapter 2: Angiotensin-(1-7)/Mas Decreases Thrombosis in Bdkrb2-/- Mice by Increasing NO and Prostacyclin to Reduce Platelet Spreading and GPVI Activation (Adopted from published paper Blood. 2013 Apr 11;121(15):3023)

INTRODUCTION

Major hypertension clinical trials show that treatment with various antihypertensive medications leads to ~20% reduction in myocardial infarction, stroke, and the need for coronary revascularization42,121. The mechanisms for the observed arterial thromboprotection are not completely known. Regulation of hypertension reduces shear and vascular injury. Other mechanism(s) through the renin angiotensin system (RAS) influence arterial thrombosis risk as well.

Previously we observed that mice (Bdkrb2-/-) lacking the bradykinin B2 receptor

(B2R) are protected from arterial thrombosis by a paradoxical mechanism that includes increased plasma angiotensin II (AngII) and renal angiotensin receptor 2

(AT2R) 40. In the Bdkrb2-/-, there is increased angiotensin converting enzyme

(ACE or kininase II) activity that converts angiotensin I to AngII 40,122. The evidence for increased ACE activity is the finding that Bdkrb2-/- plasma has elevated bradykinin 1-5, the ACE breakdown product of bradykinin (BK) 12,40,123.

Bdkrb2-/- also have increased AT2R mRNA and protein 40,124. Since AngII binds to the angiotensin receptor 1 or AT2R with equal affinity, the receptor more highly expressed determines the dominant phenotype. AngII binding to the AT2R increases nitric oxide (NO) and PGI2 17,40. Bdkrb2-/- mice have long tail bleeding times presumably due to the increased plasma NO and PGI2 40. If the AT2R, NO synthase, or cyclooxygenase 2 is blocked with a specific inhibitor, respectively, the prolonged thrombosis and bleeding times in Bdkrb2-/- normalize 40. Since

Bdkrb2-/- are not constitutively hypertensive, understanding the pathways that

32 protect them from thrombosis indicate mechanisms for arterial thrombosis risk modulation independent of blood pressure lowering 125.

New investigations indicate that the lowering of plasma AngII alone is not sufficient to correct the thrombosis protection in Bdkrb2-/-. This finding suggests an additional thromboprotective mechanism in Bdkrb2-/-. Further, the mechanism(s) for the long bleeding time in Bdkrb2-/- has not been elucidated 40.

The present investigation indicates that a second receptor of the RAS, Mas, contributes to thrombosis protection in Bdkrb2-/- also through elevation of plasma

NO and PGI2. Additionally, platelet inhibition and thrombosis protection in

Bdkrb2-/- is produced in part by acquired defects in integrin-mediated spreading and glycoprotein VI (GPVI) activation. Since the B2R, AT2R and Mas are vascular and renal receptors, modulation of these components has the potential to influence arterial thrombosis risk 52,126,127.

RESULTS

Characterization of angiotensin-(1-7) and Mas of the renin angiotensin system in

Bdkrb2-/- mice.

In our previous investigation, we observed that an ACE inhibitor, ramipril, and an antagonist to the AT2R, PD123319, independently corrected the lengthened thrombosis time in Bdkrb2-/- mice 40. As a negative control, we examined if losartan, an angiotensin receptor 1 antagonist, had any influence. As expected, losartan had no influence on the time to thrombosis on the Rose Bengal assay in

Bdkrb2-/- (61±5 min, n=5,versus untreated Bdkrb2-/- 71±3 min, n=10) (p=0.13)

(Figure 5A). However to our surprise, plasma AngII levels of losartan-treated

Bdkrb2-/- (40±30 pg/ml, n=4) were significantly lower (p<0.022) than untreated

Bdkrb2-/- mice (258±144 pg/ml, n=5) (Figure 5B). Losartan treatment did not lower AngII levels in wild type mice (45±6 pg/ml, n=4 versus untreated mice

40±20 pg/ml, n=5)(p=0.61) (Figure 5B). Our previous understanding of the mechanism for thrombosis protection in Bdkrb2-/- mice required elevated serum

AngII and its receptor AT2R 40. These unexpected findings with losartan suggested that an additional, previously unappreciated thromboprotective mechanism(s) was operative 40.

Since losartan treatment in vivo is known to increase BK levels by reduced ACE metabolism in humans, we determined renal ACE mRNA levels in our animals

128. Losartan treatment significantly decreased renal ACE mRNA in both wild type

(B2R+/+) mice (p=0.03) and in Bdkrb2-/- (p=0.017) (Figure 5C). The effect of losartan on renal ACE mRNA is a drug-class specific effect; treatment of Bdkrb2-

/- mice with telmisartan, another angiotensin receptor 1 antagonist, also lowered renal ACE mRNA 3-fold (Figure 5D). These data suggested that losartan treatment lowered AngII levels in Bdkrb2-/- by reducing ACE mRNA.

Since ACE inhibition lowers AngII levels and corrects thrombosis protection in

Bdkrb2-/- 40, and angiotensin receptor 1 antagonists lower AngII levels but do not correct thrombosis delay, an additional mechanism(s) for thromboprotection mediated by ACE was sought. Mas is the receptor for angiotensin-(1-7) [Ang-(1-

7)], a metabolic product of AngII 127. Stimulation of Mas, like the AT2R, results in vasodilation and increased PGI2 and NO formation 129-131. We determined the

35 levels of Ang-(1-7) in Bdkrb2-/- mice. In the absence or presence of losartan,

Bdkrb2-/- mice (18.4±2 pg/ml, n=8 for untreated and 20.4±1.9 pg/ml, n=8 for treated) had significantly increased plasma Ang-(1-7) over Bdkrb2+/+ animals

(14±0.1 pg/ml, n=16 for untreated and 12.7±1.6 pg/ml, n=8 for treated) (p<0.05)

(Figure 6A). Losartan treatment had no influence on plasma Ang-(1-7) levels. In addition, we next examined if there were changes in the major enzymes, angiotensin converting enzyme 2 (ACE2) and prolylcarboxypeptidase (PRCP), that produce Ang-(1-7) from AngII 132. Losartan also did not influence ACE2 or

PRCP renal mRNA in Bdkrb2-/- mice, although it increased PRCP mRNA in

Bdkrb2+/+ kidneys compared to untreated mice (Figure 6B & 6C).

To further characterize the source of constitutively elevated Ang-(1-7) in Bdkrb2-/- mice, we examined the ACE2 and PRCP activity in kidneys from untreated or losartan-treated Bdkrb2+/+ and Bdkrb2-/- mice using 3 different assays:

36 fluorogenic (Figure 7A & 7B), in vitro Matrix-assisted laser desorption/ionization

(MALDI) (Figure 7C) and MALDI imaging (Figure 7D & 7E). With each assay, there is no significant difference in ACE2 or PRCP enzymatic activity in regards to Ang-(1-7) formation in Bdkrb2-/- kidney in the absence or presence of losartan.

In summary, losartan treatment does not influence ACE2 or PRCP either at the mRNA or enzymatic activity levels in Bdkrb2-/- animals. The observed levels of plasma Ang-(1-7) in Bdkrb2-/- are independent of plasma or tissue AngII, ACE,

ACE2, and PRCP. The increased amount in Ang-(1-7) levels in Bdkrb2-/- mice may be due to reduced degradation.

Ang-(1-7) has been proposed as a mediator through which captopril and losartan contribute to thrombosis protection 51. Investigations next sought to determine if the Ang-(1-7) receptor Mas participated in the thrombosis protection seen in

Bdkrb2-/- mice. Studies showed that Bdkrb2-/- have ~2-fold increased renal Mas mRNA (p<0.017) over wild type mice (Figure 8A). This result was independent of losartan treatment. Immunoblot studies indicated that there was increased Mas antigen in the Bdkrb2-/- vs wild type, similar to what we had reported for the AT2R

(Figures 8B & 8C) 40. These combined studies suggest that Mas could be an additional receptor contributing to the thrombosis protection in Bdkrb2-/- mice.

Influence of Mas on thrombosis protection in Bdkrb2-/- mice.

Previous studies showed that treatment of the Bdkrb2-/- mice with an AT2R antagonist, PD123319, alone normalized their thrombosis protection 40 (Figure

9). We now determined if treating Bdkrb2-/- mice with a Mas antagonist, A-779, alone also reduced thrombosis times.

In wild type mice, A-779 treatment shortened the occlusion time from 34±1.7 min

(n=6) to 27±1.8 min (n=6) (p<0.02) (Figure 10A). In Bdkrb2-/-, A-779 shortened the thrombosis times from 58±2 min (n=5) to 38±2 min (n=4) (p<0.0001). When wild type or Bdkrb2-/- mice were treated with both A-779 and PD123319, there was no further shortening of the occlusion times (Figure 10A). Likewise, A-779 treatment of Bdkrb2-/- also shortened the tail bleeding time from 170±5 sec (n=6) to 88±3 sec (n=6) (p<0.0001) (Figure 10B). Mas contributed to plasma NO and

PGI2 levels. As previously reported, Bdkrb2-/- had 2-fold increased (p<0.0005)

39 plasma nitrate (21.5±1.5 M, n=6) compared to wild type mice (10±2 M, n=6)

(Figure 10C). When the Bdkrb2-/- were treated with A-779, the plasma nitrate level fell to 15±2 M, n=5, a value significantly less (p<0.05) than Bdkrb2-/- mice, but not significantly greater than wild type (p=0.092) (Figure 10C). Likewise,

-/- Bdkrb2 had significantly elevated (p<0.004) 6-keto-PGF1 levels (259±42 pg/ml, n=6) compared to wild type mice (88±18 pg/mg, n=6) (Figure 10D) 40. When

-/- Bdkrb2 were treated with A-779, the plasma 6-keto-PGF1 value fell to 132±26 pg/ml, n=5, a value not significantly different from wild type (Figure 10D). These combined studies indicated that Mas was an independent regulator of arterial thrombosis risk in Bdkrb2-/-.

Platelet function of Bdkrb2-/- mice.

Although the receptors AT2R and Mas influence plasma NO and PGI2 levels in

Bdkrb2-/-, the precise mechanism(s) for thrombosis protection in these animals is not known. We asked if increased plasma NO and PGI2 altered platelet function in the Bdkrb2-/-. When adhered to collagen, Bdkrb2-/- platelets were observed to have 1.5-fold increased DAF-FM fluorescence, a marker for intracellular NO

(Figure 11A). Additionally, resting washed Bdkrb2-/- platelets (n=6) were observed to have slightly increased (p<0.05) cGMP levels [3.7±0.5 pmol/108 platelets (n=6) vs 3.1±0.5 pmol/108 platelets in wild type (n=6)] (Figure 11B).

Further, when platelets were prepared with the phosphodiesterase inhibitor

IBMX, Bdkrb2-/- platelets trended towards increased cAMP levels [17±5 pmol/108 platelets (n=12) vs 9±2 pmol/108 platelets in wild type (n=10)] that were not statistically significant (p=0.14) (Figure 11B)). Since IBMX elevates platelet cAMP, we also prepared platelets treated with aspirin to inhibit internal prostaglandin synthesis 133-135. Aspirinated Bdkrb2-/- platelets also had elevated

(p=0.13) cAMP [15.8±4 pmol/108 platelets (n=6) vs 9±2 pmol/108 platelets in aspirinated wild type (n=6) (p=0.13)]. These combined data suggested that resting Bdkrb2-/- platelets constitutively have slightly increased cGMP and cAMP levels.

Investigations next examined if there were any platelet function defects. Platelet aggregation studies in PRP revealed that the minimal concentration that produced maximal aggregation for ADP- [20±0.4 M for wild type (n=4) vs 25±6

M for Bdkrb2-/- (n=7) platelets] or -thrombin- [97±16 nM for wild type (n=8) vs

96±7 nM for Bdkrb2-/- (n=10) platelets] were not significantly different. On flow cytometry, ADP-induced fibrinogen binding of washed platelets was not significantly different between Bdkrb2-/- platelets and wild type (Figure 12).

In additional studies, -thrombin (0.01-3 nM) or -thrombin (2-100 nM) produced the same degree of activation of integrin 2b 3 as determined by the JON/A antibody or P-selection expression on platelets from Bdkrb2-/- and wild type

(Figure 13). The EC50 for - (~1.8 nM) and - (~14 nM) thrombin-induced integrin activation or P-selectin expression were similar for Bdkrb2-/- and wild type platelets (Figure 13). These data indicated that Bdkrb2-/- platelets have no defect to ADP- or thrombin-induced activation.

Bdkrb2-/- platelet spreading

Bdkrb2-/- platelets were observed to adhere similarly to collagen as wild type.

Using different concentrations of platelets (Figure 14A) or with different incubation time (Figure 14B), Bdkrb2-/- and B2R+/+ platelets showed similar degree of static adhesion on collagen

However, Bdkrb2-/- platelets (0.64±0.1 Relative Spreading, n=30 fields from 3 independent experiments) were noted to have a 36% reduction in spreading area as determined from pixels analyzed by ImageJ compared to control platelets

(1.0±0.1 Relative Spreading, n=28) (p<0.0001) when adhered to collagen

(Figure 15A & 15B). We next determined if exogenous NO donor or PGI2 analog induced a spreading defect on collagen in wild type platelets. In these experiments, we mimicked the defect in Bdkrb2-/- platelets observed ex vivo. Wild type washed platelets were incubated with 1-100 M DEA NONOate, a NO donor, or carbaprostacyclin (100-900 ng/ml) followed by centrifugation and re- suspension in buffer without the inhibitor. Reduced spreading on collagen was observed in wild type platelets with carbaprostacyclin treatment (100-900 ng/ml) but not DEA NONOate (up to 100 M) (Figure 15C & 15D). When Bdkrb2-/- mice were treated in vivo with the combined antagonists, L-NAME and nimesulide, inhibitors of eNOS and cyclooxygenase 2, respectively, the spreading defect corrected (0.95±0.11 Relative Spreading, n=19) (p<0.0001) (Figure 15A & 15B).

Similarly, when Bdkrb2-/- mice were treated in vivo with A-779, the spreading

44 defect also corrected (0.88±0.1 Relative Spreading, n=13) (p<0.001) (Figure

15A & 15B).

Further studies showed that Bdkrb2-/- platelets have a spreading defect on the collagen peptide GFOGER that recognizes the integrin 2 1 [0.46±0.01 Relative

Spreading (n=12) in Bdkrb2-/- platelets versus 1.001±0.03 Relative Spreading

(n=12) in control platelets (p<0.0001)] (Figure 16A & 16B) 84. In vitro studies showed that treating wild type platelets with 100 M DEA NONOate (100 M) or carbaprostacyclin (300-900 ng/ml) also induced a spreading defect on GFOGER

(Figure 16C & 16D).

Bdkrb2-/- platelets also were observed to have reduced spreading on fibrinogen that recognizes the integrin 2b 3, but normal spreading on collagen related peptides (CRP) that recognize GPVI (Figure 17). Finally, if washed Bdkrb2-/- platelets were incubated for 2 h at room temperature, the spreading defect on collagen resolved (Figure 18). These combined data indicated that the elevation of plasma NO and PGI2 in Bdkrb2-/- mice produced an acquired platelet spreading defect that is mediated by integrins and reversible over time.

GPVI activation in Bdkrb2-/- platelets

Because elevated cAMP and cGMP mediated by NO and PGI2 inhibit GPVI dimerization, we examined if Bdkrb2-/- platelets also have altered CRP- and

136 convulxin-induced platelet activation . CRP-induced 2b 3 integrin activation

(JON/A binding) and P-selectin expression in Bdkrb2-/- platelets were significantly reduced (Figures 19A & 19B). The EC50 for CRP for integrin activation and P- selectin expression was 42- and 212-fold higher, respectively, in Bdkrb2-/- than wild type platelets. When convulxin was used as an agonist, Bdkrb2-/- platelets also had reduced integrin activation and P-selectin expression (Figures 19A &

19B).

In independent studies we found that wild type and Bdkrb2-/- mice have equal amount of total GPVI antigen in platelet lysates on immunoblot. However on flow cytometry, membrane-expressed GPVI antigen was decreased by 31% on

Bdkrb2-/- platelets (Figure 20). Reduced GPVI membrane expression but equal total GPVI on Bdkrb2-/- platelets may be sign of reduced ability to activate these platelets 137.

Since a reduction in the number of GPVI epitopes might account for reduced activation by CRP or convulxin, we determined if in vivo treatment of Bdkrb2-/- mice with the Mas antagonist A-779 corrected CRP-induced platelet activation.

As shown in Figures 21A and 21B, in vivo treatment of Bdkrb2-/- mice with A-

779 corrected both 0.6 and 1 g/ml CRP-induced P-selectin expression defect.

These data indicated that systemic Mas receptor over-expression in part resulted in the CRP activation defect in Bdkrb2-/- platelets.

We next determined if in vitro treatment with either NO or PGI2 alone contributed to the CRP activation defect observed in Bdkrb2-/- platelets. Washed platelets pretreated with increasing DEA NONOate (0.1 to 100 M) did not block 0.3 g/ml

CRP-induced integrin activation or P-selectin expression (Figure 22A & 22B).

When Bdkrb2-/- mice were treated with L-NAME, in vivo, CRP-induced P-selectin expression normalized only at 1 g/ml (Figure 22C & 22D).

Alternatively washed wild type platelets pretreated with carbaprostacyclin (300 to

1200 ng/ml) had significantly decreased CRP-induced (0.3 g/ml) integrin activation and P-selectin expression (Figure 23A & 23B). In order to determine if inhibition of PGI2 corrected the CRP-induced platelet defect in Bdkrb2-/- platelets, the mice were treated with nimesulide (Figure 23C & 23D). In vivo nimesulide treatment alone was able to partially correct the integrin activation and P-selectin expression defect induced by 1 g/ml CRP and P-selectin expression induced by

0.6 g/ml CRP in Bdkrb2-/- platelets (Figure 23C & 23D). These data indicate that the observed GPVI activation defect in Bdkrb2-/- platelets was in part due to elevation of PGI2 and, to a lesser extent, NO.

It has been recently recognized that GPVI dimerization requires the lowering of intraplatelet cAMP 98,136. Constitutive dimer formation is essential for GPVI activity 138. Elevation of cAMP by forskolin, prostaglandin E1 (PGE1), or PGI2 inhibits GPVI dimer formation and blocks its subsequent signaling events. Upon

GPVI activation, Syk is activated through a signaling cascade involving Src family kinases and Src homology (SH) 2 domain-containing leukocyte phosphoprotein of 76 kDa (Slp-76) (Figure 3) 70. We next examined ligation dependent Syk phosphorylation (Tyr519/520) in Bdkrb2-/- platelets. Using different concentrations

(1-10 nM) of convulxin, we found Bdkrb2-/- platelets have reduced pSyk compared to wild-type platelets (Figure 24A). Additional data showed that

51 reduced tyrosine phosphorylation of Syk in Bdkrb2-/- platelets is recapitulated in

WT (Bdkrb2+/+) platelets treated with carbaprostacyclin. Using convulxin, we found that there is significantly reduced Syk phosphorylation (Tyr519/520) in wild-type platelets treated with carbaprostacyclin (10-1000 ng/ml) compared to untreated platelets (Figure 24B). DEA NONOate (1-100 M) has no effect on

Syk phosphorylation. These data suggests that the GPVI defect is upstream of

Syk activation.

We next examined if the platelet function defects seen in Bdkrb2-/- mice were acquired from the host or intrinsic to the platelets. Bone marrow transplantation experiments were performed with wild type and Bdkrb2-/- animals. When wild type bone marrow was transplanted into a Bdkrb2-/- host, the collected platelets had reduced spreading on collagen similar to that observed when Bdkrb2-/- bone marrow is transplanted into a Bdkrb2-/- host (Figure 25A & 25B). Alternatively, when Bdkrb2-/- bone marrow is transplanted into a wild type host, the collected platelets spread on collagen similar to a wild type bone marrow transplanted into

52 wild type mice (Figure 25A & 25B). Next, we determined if bone marrow transplantation altered the thrombosis phenotype of the host animal. Carotid artery vessel closure times of Bdkrb2-/- mice transplanted with wild type bone marrow were not significantly different from those observed when a Bdkrb2-/- host received Bdkrb2-/- bone marrow (Figure 25C). Likewise the vessel occlusion times for wild type mice transplanted with Bdkrb2-/- bone marrow were the same as a wild type host transplanted with wild type bone marrow (Figure 25C). These combined studies indicated that the platelet and thrombosis phenotype observed in Bdkrb2-/- mice derived from the host and were not due to an intrinsic platelet defect.

DISCUSSION

This investigation shows that the renin-angiotensin system receptor Mas modulates arterial thrombosis potential in the intravascular compartment. Like the AT2R, increased Mas compensates for the absence of the B2R contributing to elevated intravascular NO and PGI2 17,40,124,129,130. In vitro and ex vivo studies suggest that elevated plasma PGI2 interferes with platelet activation better than

NO, producing a spreading and CRP- or convulxin-induced activation defects.

These acquired platelet function defects contribute to the delayed carotid artery thrombosis times on the Rose Bengal model. The pathways described here are important to understand how use of common antihypertensive medications like

ACE inhibitors or angiotensin receptor 1 antagonists lead to a 20% reduction in arterial thrombosis such as myocardial infarction and stroke 42,121. Further, these studies indicate how subtle platelet defects produced by changes in NO and

PGI2 alter arterial thrombosis risk in vivo.

The finding that losartan treatment of Bdkrb2-/- mice reduced plasma AngII without correcting the thrombosis delay challenged our previous interpretation that thrombosis protection in these mice was due to the double finding of elevated AngII and overexpression of the AT2R (Figures 5A & 5B) 40. Alternative explanations were needed. Losartan and its class-related agent telmisartan, in addition to angiotensin 1 receptor antagonism, decreases ACE mRNA, suggesting that this mechanism produced the reduced plasma AngII levels

(Figure 5C & 5D). Acute administration of losartan elevates AngII, whereas steady-state treatment is associated with reduced plasma AngII levels 139,140.

Previously we showed that ACE inhibition lowers AngII levels and corrects the time to thrombosis 40. Even though losartan lowers ACE, there was no decrease in plasma Ang-(1-7). In fact, in the Bdkrb2-/- mice, Ang-(1-7) is slightly increased in the absence or presence of losartan (Figure 6A). Our studies show that losartan does not alter the production of Ang-(1-7) by its two major forming enzymes, ACE2 and PRCP (Figure 7) 132. However it is presently unknown if the slightly elevated Ang-(1-7) levels in Bdkrb2-/- mice is due to reduced clearance.

Using losartan only indicated to us that an additional agent influenced by ACE is also contributing to thrombosis protection. Ang-(1-7) is recognized to have an antithrombotic effect 51-54. Ang-(1-7) mediates its effect through Mas 127. Our previous studies reported no increase in renal Mas but a 1.62-fold increase in liver Mas in Bdkrb2-/- mice 40. We re-examined renal Mas mRNA levels and presently found a 2-fold increase in Bdkrb2-/- mice that is not influenced by losartan treatment (Figure 8A). These present findings are validated by additional immunoblot studies indicating increased Mas along with AT2R antigen in Bdkrb2-/- kidneys as previously reported 40. Importantly, Ang-(1-7) and its activation of Mas is the candidate second mechanism for thrombosis protection in Bdkrb2-/- mice because Ang-(1-7) levels do not fall with losartan treatment even though AngII levels do.

The Mas receptor and its agonist Ang-(1-7) have been recognized to have an antithrombotic effect. Mas KO mice have increased venous thrombus size and short bleeding times 52. Activation of ACE to produce more Ang-(1-7) or administration of an orally active form of Ang 1-7 produces a Mas-dependent

55 antithrombotic effect in rats 53,54. We confirmed that Mas contributes to thrombosis protection seen in Bdkrb2-/- mice because systemic treatment with the

Mas antagonist A-779 shortens the time to arterial thrombosis, shortens the tail bleeding time, and lowers plasma nitrate and 6-keto-PGF1 . The ability of the

Mas antagonist A-779 to correct the thrombosis phenotype of the Bdkrb2-/- mice is identical to that observed when AT2R antagonist PD123319 is used 40. These results suggest that in the absence of the B2R, both Mas and the AT2R become over-expressed and produce increased NO and PGI2 to compensate. However, neither receptor alone fully compensates for the loss of the B2R and inhibition of either is sufficient to correct the prolonged thrombosis and bleeding time to normal.

Since the tail bleeding time is prolonged in the Bdkrb2-/- mice, we determined if there is a platelet defect. Bdkrb2-/- platelets have increased DAF-FM a marker of

NO. This finding is due to increased in vivo plasma NO or angiotensin-(1-7) stimulation of platelet Mas receptor 52. Resting Bdkrb2-/- platelets also have slightly increased cGMP and cAMP consistent with an elevation of plasma NO and PGI2 98. However, no defects in ADP- or thrombin-induced platelet aggregation were observed. Further, there were no defects in ADP-induced fibrinogen binding or - or -thrombin-induced integrin activation or P-selectin expression. Although Bdkrb2-/- platelets adhered normally to collagen, they have decreased spreading. Further, they have normal spreading on CRP but decreased spreading on fibrinogen and GFOGER, a peptide designed to

84 recognize the platelet integrin receptor for collagen, 2 1 . These data indicate

56 that there is an integrin dependent spreading defect in Bdkrb2-/- platelets. The relationship between elevated PGI2 and NO and reduced platelet spreading was evaluated in a series of in vitro and in vivo studies. In vitro treatment of wild type platelets with the synthetic prostaglandin analogue carbaprostacyclin, and to a lesser extent the NO donor DEA NONOate, creates a spreading defect on collagen and GFOGER. In vivo treatment of Bdkrb2-/- mice with the Mas antagonist A-779 or nimesulide and L-NAME, corrects the spreading defect. The mechanism(s) by which PGI2 and NO induce an integrin dependent spreading defect is presently not completely known.

In addition to the spreading defect, Bdkrb2-/- platelets have reduced CRP- and convulxin-induced integrin activation and P-selectin expression suggesting that the elevated plasma NO and PGI2 influences GPVI 70. In in vitro studies, carbaprostacyclin, but not a NO donor, induce a GPVI activation defect in wild type platelets. Moreover, in vivo treatment with A-779, nimesulide, or L-NAME partially corrects the CRP-induced platelet activation defect. Recent studies indicate that elevation of cAMP, cGMP, and PGI2 inhibit GPVI dimerization which is essential for its function 136,138. It is likely that the GPVI activation defect observed in Bdkrb2-/- platelets is related to this mechanism.

Resting Bdkrb2-/- platelets were observed to have ~30% reduction of membrane

GPVI with equal total amounts of GPVI by immunoblots of lysates. It has been shown that only 20% normal GPVI is sufficient to produce full platelet activation by collagen 141. The reduction of membrane GPVI on Bdkrb2-/- platelets alone cannot account for the spreading defect on collagen or reduced CRP-induced

57 activation. The observation that Bdkrb2-/- platelets have a spreading defect on

GFOGER peptides and fibrinogen, but normal spreading on CRP, suggest that it is independent of GPVI. Further, the fact that platelet incubation corrects the spreading defect indicates that it is an acquired defect. The bone marrow transplantation experiments confirm that the platelet spreading defect is acquired from the host. Platelets produced from transplanted bone marrow regardless of donor phenotype acquire the spreading phenotype of their host.

Recent investigations indicate that COX-2-/- mice have shortened arterial thrombosis times and deletion of vascular COX-2 is sufficient to explain their thrombosis risk 116,118. Inhibition of vascular COX-2 influences expression of eNOS and the release and function of NO 98. Most recently it has been recognized that PGI2 regulates arterial thrombus formation by suppressing tissue factor expression in vasculature, leukocytes, and microparticles 119. Although it is not precisely known if elevated PGI2 is the major contributor to thrombosis protection in our mice, the thrombosis phenotype of Bdkrb2-/- is not influenced by the phenotype of the bone marrow donor. Our data suggest that host factors, derived from vasculature in the Bdkrb2-/- mice, are the major determinant for the thrombosis protection observed.

In conclusion we have described a novel pathway by which alterations in the vascular renin-angiotensin system receptors modulate arterial thrombosis potential in the intravascular compartment (Figure 26). In the absence of the

B2R, elevated AngII or its metabolized product, Ang-(1-7), bind to overexpressed

AT2R or Mas, respectively, and increase plasma NO and PGI2. Both plasma NO

58 and PGI2 influence the platelets to produce defects in spreading mediated by integrins and GPVI activation. These pathways contribute to the reduced arterial thrombosis potential seen in Bdkrb2-/- mice. Chronic B2R blockade with an antagonist used in the management of attacks for hereditary angioedema may provide thromboprotection through this mechanism 12,40,123. Understanding these pathways are important to appreciate how common antihypertensives like ACE inhibitors or angiotensin receptor antagonists are associated with a 20% reduction in arterial thrombosis as seen in myocardial infarction and stroke. Last, appreciating the subtle differences in vascular and plasma factors observed in the present study begin to clarify the variability in arterial thrombosis risk among individuals.

Chapter 3: Angiotensin-(1-7)/Mas Protects Klkb1-/- Mice from Thrombosis by Increased Prostacyclin to Reduce Vascular Tissue Factor through Elevated Sirt1

INTRODUCTION

Prekallikrein (PK), the precursor for plasma kallikrein (KK), circulates in complex with high-molecular-weight kininogen (HK) in the plasma142. PK can be activated by enzyme factor XII (fXIIa) in the plasma to generate KK, it also can be activated independent of fXIIa on endothelial cells by prolylcarboxypeptidase

(PRCP) 143. KK is a serine protease that participates in blood coagulation during contact activation by cleaving zymogen factor XII (fXII) into fXIIa in a reciprocal manner, amplifying the autoactivation of fXII, which initiates the waterfall coagulation cascade in the intrinsic pathway leading to thrombin generation and fibrin formation. In addition to its role in contact activation, KK also promotes inflammation through the kallikrein kinin system (KKS) by cleaving HK to liberate the vasoactive peptide bradykinin (BK). BK binds to constitutively expressed B2 receptor (B2R) in the intravascular compartments to induce vasodilation and increase vascular permeability144. The PK activation and BK release is a physiological process, since deficiency of the major KK inhibitor, C1 inhibitor, causes constitutive BK mediated angioedema 144.

PK deficiency in humans (Fletcher trait) causes significantly prolonged activated partial thromboplastin time (aPTT), which can be corrected by prolonged incubation of the plasma in a glass cuvette145. PK deficient patients do not have a hemostatic defect145,146. Although PK activation promotes blood coagulation, the product of PK activation, BK, is a low affinity thrombin inhibitor and promotes blood flow by elevating nitric oxide (NO) and PGI2 and inducing tissue type plasminogen activator (tPA) formation123,126,147,148. A recent study in mice shows

61 that selective reduction of PK by antisense oligonucleotide (ASO) produces an antithrombotic phenotype without defective hemostasis149. Moreover, PK deficient mice (Klkb1-/-) are protected from ferric chloride induced arterial thrombosis and oxidative venous thrombosis with mildly prolonged tail bleeding time150.These combined studies indicate that PK deficiency is associated with thrombosis protection, not hemostasis.

The precise mechanism(s) leading to thrombosis protection in PK deficient mice is not known. This investigation indicates a novel mechanism for thrombosis risk reduction in Klkb1-/- mice independent of contact activation. We show that Klkb1-/- mice, unlike FXII deficient (F12-/-) mice, are not protected from lethal thromboembolisms in two models induced by contact activation. Klkb1-/- mice, have reduced bradykinin delivery to its B2 receptor (B2R). We observed in

Bdkrb2-/- (B2R KO) mice that defective BK/B2R signaling leads to over- expressed Mas receptor from the renin angiotensin system (RAS) producing elevated plasma PGI2 that protects them from induced arterial thrombosis40,151.

We asked if a related mechanism is occurring in Klkb1-/- mice. Our studies show that Klkb1-/- mice have elevated renal Mas and elevated plasma PGI2 that produces increased aortic expression of vasoprotective transcriptional factor sirtuin-1 (Sirt1) with decreased tissue factor (TF) mRNA. This pathway for thrombosis protection highlights the interaction between KKS and RAS in this process and the role of Mas-PGI2 axis in the modulation of arterial thrombosis risk.

RESULTS

Characterization of Klkb1-/- mice

The genotype of Klkb1-/- mice was confirmed by polymerase chain reaction

(PCR) analysis, in which tail DNA from wild type (WT, Klkb1+/+) and Klkb1-/- produced a 245 bp and 391 bp band, respectively (Figure 27A upper panel).

Immunoblot studies showed no detectable PK antigen in Klkb1-/- plasma (Figure

27A bottom panel). In addition, acid-treated Klkb1-/- plasma had less than 5% plasma kallikrein activity on a chromogenic enzymatic assay compared to normal plasma (Figure 27B). Similar to the Fletcher trait in humans, Klkb1-/- plasma had significant prolongation of aPTT. The mean clotting time in WT is 38.6±1.8 sec while the Klkb1-/- plasma did not clot up to 200 sec. The prothrombin time is not significantly different between the two genotypes (16.4±0.8 sec, n=4 in WT vs

16.9±0.6 sec, n=4 in Klkb1-/-, p>0.05) (Figure 27C). These combined data confirmed the absence of PK antigen and activity in Klkb1-/- mice. On CBC,

Klkb1-/- mice had a WBC count, hemoglobin, mean corpuscular volumes that were not significantly different than wild type mice.

Recent studies showed an antithrombotic effect in Klkb1-/- mice150. We confirmed that and expanded upon those findings. Using 4% ferric chloride, we observed pronounced thrombosis protection in Klkb1-/- mice. All WT animals had complete occlusion in carotid artery within 10 min (5.0±1.2 min, n=5), while 5 out of 7

Klkb1-/- mice did not occlude up to 60 min (p<0.005) (Figure 27D). Similarly, on the rose bengal assay, Klkb1-/- were also protected from arterial thrombosis compared to WT. The mean occlusion time in Klkb1-/- is 47.1±2.4 min (n=10),

63 which is significantly longer than the time in WT (25.0±4.6 min, n=6, p<0.005)

(Figure 27D). These combined data showed that PK deficiency had reduced risk of arterial thrombosis.

Thrombosis protection is not just defective contact activation in Klkb1-/- mice

KK participates in contact activation by cleaving fXII into fXIIa leading to thrombin generation through the intrinsic pathway. We determined the thrombin generation times (TGT) in plasma from Klkb1-/- mice. Consistent with the prolonged aPTT,

Klkb1-/- plasma had defective contact activation-induced TGT (Figure 28A). The

64 mean peak height of TGT in Klkb1-/- plasma was 37.2±6.5% (n=4) as in WT plasma (n=8), p<0.001. Upon prolonged incubation of Klkb1-/- plasma for 3 hr in a glass cuvette, the mean peak height partially corrected (72.75±4.67% as in WT, n=4), but was still significantly lower than WT plasma (n=8) (Figure 28B &28C)

(p<0.001). Similarly, the mean total area under the curve (AUC) for contact activation-induced TGT in Klkb1-/- plasma was 46.3±7.7% (n=4, p<0.001) as in

WT plasma (n=8), and it partially corrected to 75.5±3.9% (n=4) upon prolonged incubation (Figure 28B & 28D) (p<0.001).

Since Klkb1-/- plasma had reduced contact activation induced thrombin generation, we asked if defective contact activation contributed to the observed reduced thrombosis risk as seen in FXII-/- mice which are protected from collagen/epinephrine induced pulmonary embolism152. To examine the role of

65 contact activation in thrombosis protection in Klkb1-/- mice, we tested these mice in the collagen/epinephrine or long chain polyphosphate (LC polyP) models of lethal pulmonary embolism,2,152. Upon collagen/epinephrine injection, 11 out of

12 WT and 10 out 11 Klkb1-/- mice died within 30 min of challenge (survival rate

8.33% and 9.09% for WT and Klkb1-/-, respectively). Only 8 out of 12 F12-/- mice died (33.33% survival rate), which is consistent with a previous report showing survival advantage of F12-/- in this model due to defective contact activation152

(Figure 29A). When challenged with LC polyP through inferior vena cava injection, 10 out of 11 Klkb1-/- and all WT (11/11) died within 30 min of application

(Figure 29A). Alternatively, F12-/- mice had a 2 out 12 survival in the LC PolyP pulmonary embolism model2. As shown in Figure 29B, histopathology with H&E staining in the lung tissues confirmed that these mice died of massive pulmonary embolism and fibrin deposition. The number of thrombi was quantified and there was no significant difference between WT and Klkb1-/- mice in both models

(Figure 29C). In contrast, F12-/- mice had significantly lower number of thrombi compared to WT in the LC PolyP model. These combined data showed that

Klkb1-/- mice were not protected from collagen/epinephrine or LC polyP induced pulmonary embolism, indicating that the thrombosis protection seen in these animals was not just defective contact activation. Additional mechanism(s) for thrombosis protection were sought in Klkb1-/- mice.

Role of Mas receptor in thrombosis protection in Klkb1-/- mice

Since KK cleaving HK to generate BK is an important function for vascular permeability, we determined the plasma BK level in PK deficient mice. Like F12-/- and Kgn-/- mice, Klkb1-/- mice had defective BK formation with a strikingly low level in the plasma (0.05±0.01 pg/ml, n=8) compared to WT (14.03±3.58 pg/ml, n=8, p<0.005) 39,153 (Figure 30A).

Since the kallikrein kinin system (KKS) and the renin angiotensin system (RAS) interact with each other at multi-layer layers to influence thrombosis risk26,40,151, we examined if there were any alterations of the critical components in the RAS and KKS in Klkb1-/- mice. As shown in Figure 30C, the mRNA for angiotensin converting enzyme (ACE) was significantly decreased in Klkb1-/- kidneys. In addition, we found 80% reduction (p<0.01) of renal mRNA for B2R, the constitutive receptor for BK although the B1R was not significantly altered

(Figure 30B & 30C). These combined data indicated severe reduction in BK delivery to and metabolism in tissues in the Klkb1-/- mice. We recently reported that Bdkrb2-/- (B2R KO) mice were protected from arterial thrombosis by a compensatory over-expressed angiotensin receptor 2 (AT2R) and Mas receptor from the RAS producing elevated PGI2 in the plasma 40,151. Since both Klkb1-/- and Bdkrb2-/- mice have attenuated BK signaling through B2R and Klkb1-/- mice have reduced AT2R mRNA, we hypothesized that the thrombo-protective mechanism in Klkb1-/- may be mediated by Mas receptor (Figure 30C).

Our studies showed that Klkb1-/- mice had 70% increased renal Mas mRNA

(p<0.05) compared to WT mice (Figure 31A). Likewise, immunoblot studies showed over-expressed Mas antigen in the kidney tissues in Klkb1-/- (Figure

31B). We next determined if there is increased Mas ligand angiotensin-(1-7)

(Ang-(1-7)) production in Klkb1-/-. Our studies showed that renal mRNA for the

Ang-(1-7) generating enzymes ACE2 and PRCP were not significantly different between WT and Klkb1-/- mice (Figure 30C). Further studies showed that Klkb1-/- had equal enzyme activity for ACE2 and PRCP as in WT mice to generate Ang-

(1-7) (Figure 32). Ang-(1-7) level was determined to be similar between WT and

Klkb1-/- (48.29±5.85 pg/ml, n=9 and 49.08±3.60 pg/ml, n=9, respectively, p>0.05)

(Figure 31C). These combined studies suggested that the predominant component in the RAS contributing to the increased thrombosis protection in the

Klkb1-/- mice may be the compensatory overexpression of Mas.

Since an enhanced Ang-(1-7)/Mas axis has been shown to be antithrombotic 51-

54,151, we determined if over-expressed Mas receptor contributed to thrombosis protection in Klkb1-/-. In vivo treatment with A-779, a Mas antagonist, did not change the thrombosis times in WT (30.3±3.0 min for untreated, n=4 versus

26.6±3.5 min for treated, n=5, p>0.05) (Figure 31D). However, A-779 treatment significantly shortened the carotid occlusion time in Klkb1-/- from 46.0±3.7 min

(n=5) to 28.6±1.7 min (n=8) (p<0.01) (Figure 31D). These combined studies indicated that over-expressed Mas in Klkb1-/- mice contributed to thrombosis protection independent of contact activation.

Thrombosis protection is mediated by elevated plasma PGI2 and vascular Sirt1

In our previous study in Bdkrb2-/- mice, we observed that the thromboprotective effect of Mas receptor is mediated by a 3-fold elevation of plasma PGI2 measured as 6-keto-PGF1 producing an acquired platelet spreading defect on integrins and GPVI activation defect 151. Presently, Klkb1-/- mice had significantly

1.64-fold increased plasma PGI2 compared to WT as determined by 6-keto-

PGF1α (123±13 pg/ml in Klkb1-/-, n=8 versus 75±10 pg/ml in WT, n=8, p<0.05)

(Figure 33).

We next determined if there were any platelet functional defects in Klkb1-/-. Initial studies showed equal platelet counts in WT and Klkb1-/- (555±34 k/µL, n=4 versus 582±38 k/ µL, n=3, respectively, p>0.05) (Figure 34A). Additional studies showed that Klkb1-/- also had a similar tail bleeding time as in WT (455±70 sec in

Klkb1-/-, n=5 versus 387±88 sec in WT, n=6, p>0.05) (Figure 34B). Unlike

Bdkrb2-/-, Klkb1-/- mice also did not have a defective GPVI activation or platelet spreading defect on collagen. As shown in Figure 34, WT and Klkb1-/- platelets had equal response to CRP (0.3 to 3 µg/ml) on flow cytometry in integrin α2bβ3 activation as detected by JON/A antibody and P-selectin expression as detected by antibody Wug.E9.

In addition, α-thrombin (0.03 to 0.3 nM) induced JON/A binding or P-selectin expression was not significantly different between WT and Klkb1-/- (Figure 34C &

34D). Phalloidin staining showed that Klkb1-/- platelets had normal spreading on collagen compared to WT as determined from pixel areas analyzed by ImageJ

(Figure 34E). Likewise, ADP (0.3 to 3 µM) induced fibrinogen binding was also similar between the two genotypes (Figure 34F). These combined data showed that Klkb1-/- platelets had a normal response to major agonists as examined in vitro. Other mechanism(s) for thrombosis protection mediated by PGI2 were examined in Klkb1-/- mice.

Since in Klkb1-/- mice, there was only a 1.64-fold increase in plasma PGI2 versus the 3-fold increase previously seen in Bdkrb2-/- mice, we asked if the PGI2 was mostly influencing vascular function and not platelet function. Recently it has been recognized that COX-2 derived PGI2 suppressed tissue factor (TF) expression through the NAD+-dependent class III histone deacetylase sirtuin-1

(Sirt1) 119,120,154. We determined that aorta from Klkb1-/- mice have elevated Sirt1 mRNA (Figures 35A). Additionally, the vasculoprotective transcription factor

Kruppel like factor-4 (KLF4) was also elevated in Klkb1-/- aorta (Figure 35A). We previously observed that Bdkrb2-/- mice also had constitutively elevated PGI2 in the plasma (Figure 10D), as cooperative evidence, we also found that aortic tissues from Bdkrb2-/- mice had elevated mRNA for Sirt1 and KLF4 (Figure 35B).

We next determined if in vitro treatment of cultured endothelial cells with PGI2 would recapitulate our observations in Klkb1-/- and Bdkrb2-/- aorta. Mouse cardiac endothelial cells (MCECs) were treated with carbaprostacyclin (cPGI2) (5 µg/ml),

72 a PGI2 analog. Carbaprostacyclin-treated cells had increased Sirt1 and KLF4 mRNA compared to untreated control (p<0.05) (Figure 35C). In addition, they also had increased KLF4 antigen as determined by immunoblot, although Sirt1 antigen was not significantly increased by carbaprostacyclin treatment (Figure

35D & 35E).

Consistent with the elevated Sirt1 expression in the aorta, Klkb1-/- aorta were observed to have decreased TF mRNA (Figure 36A). When we examined thrombin generation in Klkb1-/- plasma using low dose of TF (0.5 pM), it was constitutively lower in Klkb1-/- plasma than in normal plasma (Figure 36B). The peak height and area under the curve were both significantly lower in Klkb1-/-

73 compared to normal plasma (Figure 36C). Further, we examined endogenous

TGT without adding exogenous TF. In this experiment, mouse plasma was collected into 2 nM rHA-infestin-4, a XIIa inhibitor to block contact activation155. In the presence of absence of rHA-infestin-4, Klkb1-/- plasma were observed to have little contact activation and persistent reduced TGT compared to Klkb1+/+ plasma (Figure.36D). As quantified in Figure 36E & 36F, the mean peak height in infestin-4 treated Klkb1-/- plasma was 0.91±0.28 (n=6) versus 4.38±0.56 in infestin-4 treated WT (n=6) (p<0.05) (Figure 36E). The mean total area under the curve (AUC) in infestin-4 treated Klkb1-/- plasma was 973±381 (n=6) compared to 5359±678 in infestin-4 treated WT (n=6) (p<0.05) (Figure 36F)

These combined studies showed that Klkb1-/- mice were protected from thrombosis by a novel mechanism that is mediated by elevated plasma prostacyclin and suppressed TF expression through increased Sirt1 in the vasculature.

DISCUSSION

This investigation, together with our previous observations in Bdkrb2-/- mice, has established the Ang-(1-7)/Mas axis as an important regulator for arterial thrombosis risk. In both animals, defective BK delivery to the B2R induces a compensatory upregulation of the Mas receptor to produce elevated PGI2 in the plasma (Figure 37). In Bdkrb2-/-, elevated PGI2 induces an acquired platelet spreading defect on integrins and a GPVI activation defect to protect the mice from arterial thrombosis 151. In contrast, the constitutively increased plasma PGI2 in Klkb1-/- has a vasoprotective effect by suppressing TF expression through elevated Sirt1. These combined studies demonstrate that the antithrombotic effect of the Ang-(1-7)/Mas/PGI2 axis involves both platelet inhibition and vascular protection (Figure 37). Understanding the pathways described here is important for appreciation of the nonconventional mechanisms by which chronic administration of plasma kallikrein inhibitors in the treatment of hereditary angioedema leads to thrombosis protection 149,156,157. Our studies begin to clarify how subtle differences in the constitutive elements in the vasculature and plasma have the potential to influence the risk of thrombosis.

The observation that Klkb1-/- mice were not protected from collagen/epinephrine or LC PolyP induced lethal pulmonary embolism suggests that the mechanism for thrombosis protection in these animals is not just defective contact activation.

Recent studies indicate that the intrinsic pathway of coagulation is a target for antithrombotic treatment without compromising normal hemostasis158,159.

Selective depletion of FXII, PK or HK protects mice from induced arterial thrombosis and ischemic stroke38,149,150,152,160. However, our studies suggest that

PK does not have a critical in vivo role for contact activation induced thrombosis.

The observation that prolonged aPTT in PK deficient plasma auto-corrected upon

77 prolonged incubation with glass suggests that the role of PK in plasma clotting could be bypassed142. During in vivo thrombus formation, FXII activation could be initiated and enhanced by directly binding to collagen. However, HK/PK did not influence the binding affinity, suggesting this process is PK independent161. In addition, the equal number of thrombi in LC polyP challenged Klkb1-/- and WT indicates normal thrombin generation from contact activation in vivo in the absence of PK (Figure .29C). These results triggered our investigation looking for a contact activation independent mechanism for thrombosis protection in

Klkb1-/-.

Beside its role in FXII activation, plasma kallikrein is also the enzyme producing

BK by cleaving HK. Consistent with this premise, we observed markedly reduced

BK formation in Klkb1-/- plasma (Figure 30A). Interestingly, the constitutive receptor for BK, B2R was also significantly decreased in the Klkb1-/- kidney

(Figure 30B). This observation was similar to our recent findings in Bdkrb2-/- mice, because both animals have defective BK delivery to the target receptor

B2R in the kidneys151. Since Mas mRNA and protein were also increased in

Klkb1-/-, we hypothesized that these two animals may share a similar mechanism for thrombosis protection through Mas independent of contact activation. The finding that Mas antagonist A-779 corrected the thrombosis time in Klkb1-/- confirmed our hypothesis (Figure 31D). In both animals, Mas was over- expressed to compensate the defective BK signaling through B2R. Indeed, the kallikrein kinin system and the renin angiotensin system counter-balance with each other at multiple layers to influence thrombosis risk151. Plasma kallikrein,

78 the enzyme that produces bradykinin, activates prorenin to renin, which converts angiotensinogen to angiotensin I142. Prolylcarboxypeptidase (PRCP), an AngII degrading enzyme, activates PK independent of FXII143. Presently it is not known by what mechanism Mas was over-expressed to compensate the defective

BK/B2R signaling in these two animals. One possibility is that B2R and Mas form heterodimers. Such phenomenon has been observed among other 7- transmembrane G-protein-coupled receptors, which interact with each other at a functional level in the form of homodimers or heterodimers in RAS and KKS162.

Our studies provide a new interpretation for risk of arterial thrombosis. Under physiological conditions, where there is no collagen exposure or bacterial LC

PolyP, the KKS has the potential to influence thrombosis risk through RAS independent of contact activation. Angiotensin-(1-7)/Mas, which recently has been recognized to be antithrombotic52-54, is a key regulator during this process.

Our studies in Klkb1-/- along with the previous investigation in Bdkrb2-/- suggests that PGI2 is major mediator for the antithrombotic effect conferred by over- expressed Mas receptor in these two animals. In Bdkrb2-/-, elevated PGI2 produces an acquired platelet spreading defect in vitro, which is reversible over prolonged incubation (Figure 18)151. However, we did not observe an in vitro function defect in Klkb1-/- platelets. This difference may be due to the different extent of elevation in plasma PGI2 levels as seen in these two animals. In Klkb1-/-

, there is 1.6-fold increase of plasma PGI2 level as detected by 6-keto-PGF1α, however, in Bdkrb2-/- there is 2.9-fold increase of plasma PGI2. Our studies reveal how subtle differences in the constitutive level of plasma PGI2 have the

79 potential to influence arterial thrombosis through different mechanisms. Sirt1 has been recognized as an important antithrombotic and vasoprotective transcription factor. Inhibition of Sirt1 increases thrombosis risk by enhancing TF expression154. Depletion of Sirt1 is associated with increased susceptibility to particulate matter (PM) triggered lung coagulation through decreased KLF2 and thrombomodulin163. Recently, Barbieri et al reported that inhibited Sirt1 resulting from decreased plasma PGI2 level contributes to accelerated thrombus formation in COX-2 deficient mice through enhanced vascular TF expression. Our studies provide an additional in vivo model for how constitutive elevated PGI2 primes the vasculature to produce increased Sirt1, therefore decreasing thrombosis risk by suppressing vascular TF. This investigation, along with previous studies, has established an unconventional role of PGI2 in the regulation of thrombosis risk by protecting the vascular beds in addition to its role in platelet inhibition. In addition, our study for the first time showed that PGI2 has the potential to elevate KLF4 as part of its vaso-beneficial effect (Figure 35C). As cooperative evidence, we also observed increased mRNA of Sirt1 and KLF4 in the aorta in Bdkrb2-/- mice

(Figure 35B).

In conclusion, we have defined a novel mechanism for thrombosis protection in

Klkb1-/- mice. Altered balance between KKS and RAS in the intravascular compartments has the potential to influence arterial thrombosis risk independent of contact activation. In the absence of prekallikrein, there is defective BK formation and attenuated BK delivery to the B2R, which in turn results in compensatory over-expression of the Mas receptor to produce elevated PGI2 in

80 the plasma. This constitutively elevated PGI2 contributes to thrombosis protection in part by increasing vasoprotective transcription factors Sirt1 and

KLF4 to suppress vascular TF expression. These studies along with our previous investigation in Bdkrb2-/- suggest that the Mas-PGI2 axis is an important modulator of constitutive arterial thrombosis risk. It provides a new interpretation for thrombosis risk regulation independent of contact activation. Understanding these pathways is also important to appreciate how alterations in the plasma factors influence vascular components and modulate thrombosis risk.

Chapter 4: Methods and Materials

Methods for Chapter 2 (bradykinin b2 receptor deficient mice)

Materials.

Human -thrombin (3000 U/mg) and -thrombin were purchased from

Haematologic Technologies. The Mas receptor antagonist A-779 [H-Asp-Arg-Val-

Tyr-Ile-His-D-Ala-OH (D-Ala7-Angiotensin 1-7)] was purchased from Bachem.

Antibodies to the AT2R and Mas were purchased from Santa Cruz Biochemicals and Nova Biologicals LLC, respectively. DEA NONOate (diethylamine NONOate) and carbaprostacyclin were purchased from Cayman Chemical. Rat antibodies to murine P-selectin (Wug.E9), the activated epitope of murine 2b 3 integrin

(JON/A), and murine platelet GPVI (JAQ-1) were purchased from Emfret

Analytics. Convulxin was purchased from Alexis. CRP (collagen-related peptide) was generously provided by Dr. Deborah Newman, Blood Center of Wisconsin,

Milwaukee, WI. Losartan, telmisartan and IBMX [1-methyl-3-(2-methylpropyl)-7H- purine-2,6-dione] were obtained from Sigma. Peptide GFOGER was generously provided by Drs. Yunmei Wang and Daniel I Simon, Case Western Reserve

University 84.

Animals.

All animal care and experimental procedures complied with the principles of

Laboratory and Animal Care established by the National Society for Medical

Research and were approved by the Case Western Reserve University

Committee on Use and Care of Animals. All studies were performed on mice 8-

12 weeks of age. Bradykinin B2 receptor knockout mice (B2R-/-), strain name

B6/129S7-Bdkrb2tm1Jfh and their wild type, B6/129SF2/J mice (B2R+/+), originally

83 were purchased from Jackson Laboratories, Bar Harbor ME, but then mated to produce heterozygous animals from which wild type and gene-deleted colonies were re-derived 40. The B2R-/- colonies were maintained by brother/sister mating and all progeny were genotyped by PCR using DNA purified from tail biopsies.

B2R-/- from Jackson Laboratories were also backcrossed one generation against

129S1/svImJ mice to make heterozygous animals 40. The heterozygous animals were then bred to make homozygous and wild type animals and were demonstrated to have the same thromboprotection as the B6/129S7-Bdkrb2tmJfh

KO mice 40. The genotyping of the B2R-/- mice was detected with two pairs of probes: a sense probe 5’-CTTGGGTGGAGAGGCTATTC-3’ and antisense probe

5’-TGTCCTCAGCGTGTTCTTCC-3 and sense 5’-AGGTGAGATGACA-

GGAGATC-3’ and antisense 5’-GGTCCTGAACACCAACATGG-3’. Wild type gene produced a 361 bp band; the knockout animal had a 280 bp band 164.

Assays.

AngII antigen was determined as previously reported 40. Angiotensin-(1-7) was measured in the Hypertension & Vascular Research Center, Wake Forrest

University Health Science Center, Winston-Salem, NC. The stable analogue of

PGI2, 6-keto-prostaglandin F1 (6-keto-PGF1 ), and serum nitrite/nitrate were measured in mouse plasma according to the manufacturer's specifications

(Cayman Chemical) 40.

Data analysis.

All data are presented as means of at least triplicate determinations and are presented as mean±SEM unless otherwise indicated in the text. One way

ANOVA were performed for comparison among 3 or more related groups with a

Bonferroni correction. Two way ANOVA were applied to determine changes of several parameters between two groups. Significance between two groups is determined by the unpaired, non-parametric two-tailed t test. P values < 0.05 were considered significant. In each experiment, the statistical analysis used is reported in the figure legends.

Mouse carotid artery thrombosis studies.

Mice for the Rose Bengal carotid artery thrombosis models were anesthetized with 62.5 mg/kg sodium pentobarbital IP,immobilized on an operating table, and performed as previously reported 40,165. The time to vessel occlusion was determined after flow in the vessel stopped for 20 min as previously reported

40,165. Mouse tail bleeding times were performed as previously reported 40,165.

Preparation of mouse plasma and serum.

Animals were anesthetized with sodium pentobarbital IP or isoflurane and a celiotomy was performed to expose the inferior vena cava (IVC). Blood for plasma was obtained from mice in 3.2 gm% sodium citrate in a 1:9 ratio or in buffered acid-citrate-dextrose (ACD) in a 1:6 ratio by venipuncture of the IVC 40.

Plasma was prepared as previously reported 40. mRNA and studies.

B2R-/- and wild type control mice 8-12 weeks old had tissue collected for mRNA and protein studies as previously reported 40. Total RNA and cDNA preparation and real-time quantitative PCR were performed as previously reported 40. In the

85 present studies, detection of PCR products was performed with SYBR-Green

(BioRad, Richmond, CA). Primers for Mas, angiotensin converting enzyme

(ACE), angiotensin converting enzyme 2 (ACE2), and prolylcarboxypeptidase

(PRCP) are given in (Table I). The relative level of gene expression was

-∆∆Ct normalized to GAPDH by the formula 2 , where-∆∆Ct=∆ Ct gene - ∆ Ct Average where ∆ Ct gene = Ct gene - Ct GAPDH and ∆ Ct Average = Average ∆ Ct WT according to the procedure of Schmittgen et al. 40,166.

Table I. Primers for Real-time PCR in Bdkrb2-/- mice

Interventional studies in B2R-/- mice.

Wild type and B2R-/- mice were treated with losartan at 10 mg/kg/day dissolved in their drinking water prepared fresh daily for 10 days 167. Telmisartan was suspended in water and given orally in 100 l volumes by gavage at 5 mg/kg/day for 14 days 168. Wild type or B2R-/- also had an osmotic pump placed (ALZET

Model #2001, ALZET Corp., Cupertino CA) such that saline or 1.44 mg/kg/day of

A-779 would be administered for 7 days 169. B2R-/- were also treated for 10 days with both L-NAME (NG-nitro-L-arginine methyl ester) (Sigma) in the drinking water made fresh daily as previously reported 40 and nimesulide [N-(4-nitro-2- phenoxyphenyl)-methanesulfonamide] (Cayman Chemical) suspended in drinking water (10 mg/ml) and given by gavage in 400 l aliquots.

Determination of Renal Ang-(1-7) production in B2R-/- mice.

The production of renal Ang-(1-7) in B2R-/- kidneys and their wild type was determined by 3 assays: fluorogenic substrate, MALDI enzyme, and MALDI imaging enzyme assays 132,170. For the fluorogenic substrate assay, kidney homogenate (10-20 µg protein) was incubated with Mca-APK-Dnp substrate in

50 mM sodium citrate buffer pH 5 or potassium phosphate buffer pH 7 containing

5 mM ZnCl2, 150 mM NaCl2 and 10 µM lisinopril. Excitation and emission of fluorescence upon cleavage was detected at 328 nm and 393 nm, respectively.

No ACE2 activity is measured at pH 5 and no PRCP activity is detected at pH 7.0 in these assays.

For the in vitro MALDI enzyme assay, Ang-(1-7) formation from AngII was assessed as previously described with some modifications 170. Kidney homogenate (20 µg protein) was incubated at 37ºC in 50 mM sodium citrate buffer pH 5 (PRCP) or potassium phosphate buffer pH 7 (ACE2) containing 0.1 mM AngII, 2 mM PMSF and 20 µM bestatin. The reaction was stopped by acidification with TFA. Peptides were purified using a µ-C18 Ziptip (Millipore, MA,

USA). Mass spectra were obtained using an Autoflex III smartbeam MALDI

TOF/TOF instrument (Bruker Daltonics, MA, USA). The mass spectrometer was operated with positive polarity in reflectron mode. A total of 2000 laser shots were acquired randomly for each spot at a laser frequency of 100 Hz. The spectral analysis was performed with proprietary Bruker Flex Analysis software.

For the MALDI imaging enzyme assay, consecutive tissue sections were prepared from fresh frozen kidneys, incubated with 0.1 (for ACE2) or 1 mM (for

PRCP) AngII at 37°C for 5 min and 15 min, respectively, and spray coated with

MALDI matrix as previously described 170. MS images were obtained using an

Autoflex III smartbeam MALDI TOF/TOF instrument (Bruker Daltonics, MA,

USA). The mass spectrometer was operated with positive polarity in reflection mode. A total of 200 spectra were acquired for each spot in the range of m/z 500-

3000 at a laser frequency of 100 Hz. A 200 µm raster was used which was aligned to an optical image. The spectral analysis was performed with proprietary Bruker Imaging software.

Platelet studies.

Several platelet washing techniques were utilized for platelet function assays.

Platelet aggregation studies were performed on mouse blood collected into 3.2 gm% sodium citrate (ratio blood to anticoagulant, 9:1) by venipuncture of the inferior vena cava. The collected blood was mixed 1:1 with HEPES-Tyrodes buffer (10 mM HEPES, 12 mM NaHCO3, 130 mM NaCl, 5 mM D-glucose, 5 mM

KCl, 0.4 mM Na2HPO4, 1 mM MgCl2, pH 7.4) followed by centrifugation at 800 xg to prepare platelet-rich plasma (PRP). Platelet aggregation studies were performed on PRP (2 X 108/ml) 171. Mouse platelet aggregation to human - thrombin (10-100 nM) or ADP (10-60 M) was performed in a Chronolog aggregometer after introduction of the agonist. In these experiments, the minimal concentration to produce maximal aggregation (threshold) was determined.

Platelets for flow cytometry, cGMP and cAMP measurement, and collagen adhesion and spreading studies were prepared from mouse whole blood collected by IVC puncture and anticoagulated with buffered ACD (85 mM

88 trisodium citrate, 83 mM dextrose, and 21 mM citric acid, pH 6.5) at a 1:6 ratio with whole blood. PRP was prepared by centrifugation of whole blood at 2300 xg for 20 sec. After collecting the supernatant, the sample was centrifuged again at

2300 xg for 10 sec and the first supernatant was pooled with the second. Pooled

PRP aliquots were then pelleted by centrifugation at 2200 xg for 1 min per 200 μl of PRP and after removing the supernatant, the platelets were resuspended to desired platelet count with Hepes Tyrodes buffer, pH 7.4. Platelet NO determination was performed with DAF-FM (4-amino-5-methylamino-2,7- difluorofluorescein, Molecular Probes) on a collagen matrix as previously reported 52. Platelet cGMP was determined on washed platelets suspended in

Hepes-Tyrodes buffer containing 1 mM IBMX (Sigma). 100 l of WT or B2R-/- platelets (1 X108/ml) were pelleted by centrifugation at 2300 xg for 5 min in the presence of 10 mM EDTA. After lysis of the pellet with 200 l of lysis reagent, the intra-platelet cGMP was determined according to the manufacturer’s protocol

(Amerhsam cGMP Enzyme Immunoassay Biotrak System). Platelet cAMP was determined as platelet cGMP with 1 mM IBMX added to the anticoagulant and

Hepes Tyrodes buffer. Platelet cAMP was also determined without IBMX but with100 M aspirin added to the PRP followed by 30 min incubation at 37oC, centrifugation and pellet lysis with assay buffer 133-135. The intra-platelet cAMP was determined according to the manufacturer’s protocol (Cayman Chemicals).

Platelet flow cytometry was performed on washed platelets diluted to 0.5 X

8 10 /ml in HEPES-Tyrode’s buffer containing 1 mM MgCl2 and CaCl2. Twenty l of platelets were stimulated with increasing concentrations of collagen-related

89 peptide (CRP) (0.01-1 g/ml), convulxin (0.1-5 nM), human -thrombin (0.01-3 nM), human -thrombin (2-100 nM) or ADP (1-30 M). Thrombin-, CRP-, or convulxin-activated platelets were examined on flow cytometry for 2b 3 integrin activation with the JON/A antibody and P-selectin expression with the Wug.E9 antibody 54. ADP-stimulated washed platelets were examined for Alexa Fluor

488-labeled fibrinogen (Molecular Probes) binding on flow cytometry. In other studies, a FITC-labeled rat anti-mouse GPVI antibody (JAQ-1) was incubated with non-activated washed platelets from B2R-/- and wild type animals to determine the number of GPVI copies on the platelets.

Platelet adhesion assays were performed on 96 well plates coated with 100 l type-I fibrillar collagen (10 g/ml) in PBS for 16 h at 4oC 108. Washed B2R-/- platelets (0.1 to 1.2 X 108/ml) or control were incubated on coated plates for the indicated time period at 37oC. Non-adherent platelets were removed by washing with PBS. Adherent platelets were incubated with 100 l 0.1 M citrate buffer (31 mM citric acid, 5 mM sodium citrate dehydrate, pH 5.4) containing 5 mM p- nitrophenol phosphate and 0.1% Triton X-100 at room temperature for 1 h 108.

The reaction was stopped by the addition of 100 l 2 M NaOH and the absorbance was read at 405 nm. Platelet spreading was performed on a MatTek culture dish coated overnight with collagen, peptide GFOGER in the presence of

1 mM Ca2+, or CRP all at 10 g/ml and blocked with 3 mg/ml BSA. When spreading was performed with fibrinogen (100 g/ml), 1 mM Ca2+ was added to the buffer. The adhered platelets were fixed with 3% paraformaldehyde and then permeabilized with 0.1% Triton X-100 in PBS with 1 mg/ml BSA 70. Filamentous

90 actin was stained with Alexa-Fluor 568-conjugated phalloidin (25 l/ml)

(Molecular Probes) in PBS for 30 min at room temperature 70. Surface area in spread platelets was measured in pixels using ImageJ software (NIH) particle analysis and normalized to wild type control 70. Finally in certain experiments, washed wild type platelets were incubated with DEA NONOate (1-100 M) or carbaprostacyclin (100-900 ng/ml) for 10 min at 37oC followed by washing and re-suspension in Hepes Tyrodes buffer, pH7.4. After resting for 30 min, the resuspended platelets were examined for spreading and CRP-activation on flow cytometry.

Bone marrow transplantation.

Bone marrow cells from wild type and Bdkrb2-/- donor mice at least 8 weeks old were harvested from the femur and tibia of both hind limbs and collected in PBS with 2% fetal bovine serum under sterile conditions. B2R-/- or wild type recipient mice were lethally irradiated with 11 Gy to ablate bone marrow cells. The irradiated mice received 1 to 2 X 106 bone marrow cells in 400 l PBS by tail vein injection. After at least 6 weeks from transplantation, recipient mice had their platelets collected for spreading studies or were examined for thrombosis risk.

Immunoblot studies.

Kidney tissue was prepared for immunoblot as previously reported 40. Western blot of 4 different kidney lysates with equal total protein amounts (24 g) was performed in Bdkrb2-/- and wild type mice 40. The kidney samples were homogenized in 2 ml of RIPA buffer (1% NP40, 0.5% sodium deoxycholate and

0.1% SDS) and then centrifuged at 10,000 rpm for 10 min at 4oC and the supernatant was collected. The primary antibodies were rabbit anti-human antibody to AT2R (H143) or Mas both used at 0.2 g/ml. Both antibodies were detected with a goat anti-rabbit antibody conjugated with horseradish peroxidase

(HPR) (Cell Signaling) at 1:1000 dilution. The specific reactivity of the antibody with the electroblotted sample was detected with the ECL system from

Amersham Biosciences, Piscataway NJ. The immunoblots were scanned using

Scion image software to determine the band intensity and relative amounts of protein present. The western blot data for Mas and AT2R were normalized to renal kininogen detected by goat anti-kininogen heavy chain antisera at 1:5000

172 followed by a rabbit anti-goat IgG-HPR at 1:10,000 (Sigma) or GAPDH detected by a mouse anti-GAPDH at 1 g/ml (Millipore, Temecula, CA) followed by a goat anti-mouse IgG-HPR at 80 ng/ml (Jackson ImmunoResearch

Laboratories, West Grove, PA), respectively. Data are expressed as the mean±SEM of the ratio of target protein to control protein in kidney samples from

4 mice.

Methods for Chapter 3 (prekallikrein deficient mice)

Materials

Antibodies to prekallikrein, Mas, Sirt1, Klf2, KLF4 and β-tubulin were purchased as indicated in the section of immunoblot studies. Mas antagonist A-779 was obtained from Bachem. Carbaprostacyclin was from Cayman Chemicals. The insoluble high-molecular-weight sodium polyphosphate was generously provided by Dr. James Morrissey (University of Illinois, IL). Collagen related peptide (CRP) was a gift from Dr.Debra Newman (Blood Center of Wisconsin).

Animals

Prekallikrein knockout mice (Klkb1-/-) were generously obtained from Edward

Fenner of Harvard University and are back crossed to C57BL/6 background for 4 generations14. These animals were mated with wild-type (C57BL/6) to make heterozygous animals and then re-derived into KlKb1-/- and littermate wild type colonies maintained by brother/sister mating. The genotyping of Klkb1-/- mice are determined with two sets of primers: Forward 5’-

CTTCCAGGTAGCTGCTTTCTACC-3’ and Reverse 5’-

TCACCCACAACCTTCACAGAAAGG-3’ for wild-type (245 bp band), and

Forward 5’-CGCTGCTTAGGATGGTAGGAG-3’ and Reverse 5’-

GCTAGACTAGTCTAGCTAG-AGCGG-3’ for knockout (391 bp band)173.

Chromogenic assay of plasma prekallikrein

Mouse plasma was collected into 3.2 gm% sodium citrate and acid-treated with equal volume (50 l) of 1/6 M HCl for 15 min at room temperature174. 50 l of

93 phosphate buffer (0.1 M Na phosphate, pH 7.6, 0.15 M NaCl, and 1 mM EDTA) was then added into the sample followed by neutralization with 50 l of 1/6 M

NaOH. The samples (50 l) were prediluted with 350 l of assay buffer (50 mM

Tris-HCl, pH 7.9 and 0.1% polyethylene glycol). Samples from normal plasma

(100%) were serially diluted to prepare the standard curve. Prekallikrein activity was determined on a chromogenic reaction in which prepared plasma samples

(50 l) was activated by plasma prekallikrein activator (50 l, Enzyme Research

Labs) to cleave the substrate H-D-Pro-Phe-Arg-pNA (S-2302) (2mM, 50 l,

Chromogenix). The amount of prekallikrein was calculated as the slope of absorbance (405 nm) over time and presented as percentage of normal plasma samples according to the standard curve.

Assays

Plasma was collected into 3.2 gm% sodium citrate to measure bradykinin and the stable metabolite of PGI2, 6-keto-prostaglandin F1 (6-keto-PGF1 ) Bradykinin was determined by ELISA Kit provided by Bachem. 6-keto-PGF1 was measured in mouse plasma by ELISA according to the manufacturer's specifications

(Cayman Chemical) as previously described151. Angiotensin-(1-7) was measured in the Hypertension & Vascular Research Center, Wake Forest University Health

Science Center (Winston-Salem). Mouse complete blood counts were measured in 500 l of whole blood on a HEMAVET analyzer.

Coagulant times

The activated partial thromboplastin time (aPTT) was performed by adding 50 l citrated plasma was mixed with 50 l aPTT reagent (Helena Laboratories) in a glass tube and incubated for 5 min at 37ºC. The reaction was initiated by adding

50 l 35.3 mM CaCl2. The endpoint clotting time was determined visually by constantly tilting the cuvette in a 37ºC water bath. The prothrombin time was performed similarly as the aPTT except that 50 l citrated plasma was added to a glass tube containing 100 l pre-warmed PT reagent (Innovin, Siemens). After 5 min incubation, the reaction was initiated by the addition of 50 l 35.3 mM CaCl2.

Thrombin generation times

Mouse plasma collected into 3.2 gm% sodium citrate was used for thrombin generation times (TGT). In some experiments, rHA-Infestin-4 was also added with final concentration 2 nM in whole blood. Tissue factor-induced TGT was performed with 30 l plasma in the presence of 4 M phospholipids (Rossix

Phospholipid, diaPharma), ~1:2000 of recombinant tissue factor

(Innovin,Siemens), and 0.51 mM Z-Gly-Gly-Arg-AMC (Bachem)175. Contact activation-induced TGT was performed as above except APTT reagent (20% volume) (Helena Laboratories) was added instead of tissue factor. In both assays, the reactions were initiated by the addition of 16.4 mM CaCl2, final concentration. Substrate hydrolysis was measured in a NOVOstar fluorometer

(BMG-Labtech) with excitation set at 390 nm and emission at 460 nm. The TGT data were analyzed by peak height and area under curve (AUC) using PRISM4,

Graphpad and the values are normalized to wild-type controls.

Mouse arterial thrombosis assays

Rose Bengal thrombosis assay was performed as previously reported151. Briefly, mice were anesthetized with sodium pentobarbital IP. Flow rate within exposed carotid artery was monitored and the time to complete vessel occlusion was recorded after thrombosis was initiated by laser induced photochemical injury from Rose Bengal IV151. Using the Rose Bengal thrombosis assay, certain interventional studies were performed. Wild type and Klkb1-/- mice were treated for 7 days with saline or A-779 (1.44 mg/kg/day), a Mas antagonist, loaded in an osmotic pump (ALZET #2001, ALZET Corp.) as previously reported151 Klkb1-/- and controls were also treated with splitomicin, a Sirt1 inhibitor, (dissolved in

DMSO with further dilution in PBS) by intra peritoneal injection at a dose of 20 mg/kg/day for 5 days. Additionally, the Ferric Chloride thrombosis assay was also performed by applying a piece of Whatman paper (3 mm long X 1 mm wide)

151 saturated with 4% FeCl3 on top of the carotid artery for 1min . After removal of the paper, the vessel was washed with saline and time to complete occlusion was recorded. The tail bleeding time was performed as previously reported151.

Collagen/Epinephrine induced pulmonary embolism

A mixture of collagen/epinephrine was injected into the jugular vein of wild-type,

Klkb1-/- or F12-/- mice anesthetized with sodium pentobarbital I.P. in a dose at 0.8 mg of collagen and 60 g of epinephrine per kilogram of body weight152. Mice alive for more than 30 min were considered as survivors.

Polyphosphate induced pulmonary embolism

The soluble product of long chain polyphosphate (LC polyP) was prepared and quantified using the Malachite Green method according to a protocol from the

Morrissey lab 176,177. The concentration of our LC polyP preparation was determined to be 2 M in terms of phosphate monomer. The polyP was diluted in

PBS solution and slowly injected (60 g/g body weight) into the inferior vena cava of wild-type, Klkb1-/- or F12-/- mice anesthetized with sodium pentobarbital

IP2. Mice alive for more than 30 min were considered as survivors. Lungs were rapidly harvested and fixed in 4% paraformaldehyde. Tissues were cut and stained with hematoxylin and eosin (H&E) for histopathology analysis. mRNA studies

Total mRNA was isolated from homogenized mouse kidneys, pulverized aortas, or cell culture using the TriZOL (Invitrogen) chloroform method and first strand cDNA was synthesized with Superscript III reverse transcriptase (Invitrogen).

Real-time PCR was performed on iCycler IQ5 with SYBR-Green (BioRad).

Primer sequences are provided in Table II. The relative fold expression of target gene was normalized to control gene (GAPDH or 18S) with the formula as previously reported 151,166.

Cell Cultures

Mouse Cardiac Endothelial Cell lines (MCECs) were purchase from

CedarlaneLabs. Cells were cultured in DMEM/F-12 medium containing 5% Fetal

Bovine Serum. Cells were treated with carbaprostacyclin (5 µg/ml) or vehicle alone (0.1% DMSO) for 4 hr before harvested for mRNA studies or western blotting.

Table II. Primers for Real-time PCR in Klkb1-/- mice

Statistics

All data are presented as Mean ± Standard error of the mean (SEM) unless otherwise indicated. Difference between two groups was determined by unpaired

2-tailed student t test. One-way ANOVA analysis with a Bonferroni correction was used to compare 3 or more related groups. P value <0.05 was considered as significant difference.

Renal ACE2 and PRCP enzyme activity studies.

Ang-(1-7) formation from Ang II was assessed using the in vitro MALDI enzyme assay as previously described with some modifications 151,178. Kidney homogenate (20 µg protein) was incubated at 37ºC in 50 µl reaction mixture containing 50 mM sodium citrate buffer pH 5 (PRCP) or potassium phosphate buffer pH 7 (ACE2), 0.1 mM Ang II, 2 mM PMSF and 20 µM bestatin for 60 min.

The reaction was acidified with TFA and stable-isotope labeled internal standards

98 for Ang II (105.6 pmol) and Ang-(1-7) (101.8 pmol) were added (New England

Peptides). The mixture was 1:10 diluted with 90% acetonitrile containing 0.3%

TFA and 0.5 μl were spotted onto a MALDI target plate. Mass spectra were obtained using an Autoflex III smartbeam MALDI TOF/TOF instrument (Bruker

Daltonics). The mass spectrometer was operated with positive polarity in reflectron mode. A total of 2000 laser shots were acquired randomly for each spot at a laser frequency of 100 Hz. The spectral analysis was performed with proprietary Bruker Flex Analysis software.

Platelet Studies

Flowcytometry was performed with washed platelets as previously reported151.

Platelet response to ADP was examined by binding of Alexa-488 conjugated fibrinogen (Sigma Aldrich). α-thrombin or CRP induced platelet activation was detected by two-color system (Emfret) with PE conjugated JON/A antibody, which recognizes activated integrin α2bβ3, and FITC conjugated Wug.E9 antibody, which recognizes mouse P-selectin. Platelet spreading on collagen was performed as previously reported151. Spreading size was quantified as pixel area from phalloidin staining of cytoskeleton actin with Image J.

Immunoblot studies

For detection of prekallikrein antigen in plasma, mouse plasma samples were prepared with 3.2 gm% sodium citrate and diluted 100 folds with Laemmli sample buffer before loaded for western blot. The primary antibody was sheep anti- prekallikrein antibody (1:5000) from Enzyme Research and was detected with a rabbit anti-sheep secondary antibody (1:10000) conjugated with horseradish

99 peroxidase (HRP) from Santa Cruz Biotechnology, Inc.. Western blot for renal

Mas antigen was performed as previously reported 151. Mas antigen and its loading control -tubulin were detected using a polyclonal rabbit anti-Mas antibody (Nova Biologicals LLC, 1:5000) and a polyclonal rabbit anti-tubulin antibody (Cell Signaling, 1:1000), respectively, followed by an HRP conjugated goat anti-rabbit secondary antibody (Cell Signaling, 1:1000). Sirt1 antigen was detected using a rabbit antibody (Cell signaling, 1:1000) and an HRP conjugated anti-rabbit IgG (Cell Signaling, 1:1000). KLF4 antigen was detected by a polyclonal goat anti-mouse KLF4 (R&D, 1:400) antibody followed by an HRP conjugated rabbit anti-goat IgG (Santa Cruz Biotechnology, 1:2000).

100

Chapter 5: Summary and Future Directions

101

In sum, we have identified novel mechanisms for thrombosis protection in

Bdkrb2-/- and Klkb1-/- mice. In both animals there is attenuated bradykinin delivery to B2R receptor. The defective bradykinin signaling in vivo results in compensatory overexpression of renal Mas receptor leading to increased PGI2 production in the plasma. Elevated plasma PGI2 reduces thrombosis risk in these animals through two different mechanisms: in Bdkrb2-/- mice, constitutively elevated PGI2 produces an acquired integrin-dependent spreading defect and

-/- GPVI activation defect in platelets; in Klkb1 mice, increased plasma PGI2 reduces tissue factor expression through elevated Sirt1 in the vasculature. This study demonstrates that alterations in the components of the KKS have the potential to influence thrombosis risk through the RAS independent of contact activation. It highlights the role of Angiotensin-(1-7)/Mas pathway as an important nonconventional regulator for thrombosis risk. Future research would be promising along the following directions:

Mechanisms for the compensatory effect of Angiotensin/Mas axis on

Bradykinin/B2R signaling

It has been well established that the KKS and RAS interacts with each other at multiple levels26. Our study suggests that this interaction has the potential to influence thrombosis risk. However, it remained unclear by what mechanisms

Bradykinin signaling influences Mas receptor expression. Several in vitro approaches can be used to recapitulate our observation in vivo. These strategies are based on two different hypothesis: (1) The downstream signaling pathway of

B2R negatively regulates the gene transcription of Mas receptor. One study by

102

Saifudeen et al showed that the promoter region of B2R was cooperatively controlled by transcription factors p53, KLF4 and cAMP response element binding protein (CREB) during nephron differentiation179. Similarly, using bioinformatics approaches, the promoter region and binding proteins can be predicted for Mas receptor. It would be critical to identify if any of these candidates are downstream of B2R signaling pathway. (2) The B2R and Mas receptors form heterodimers that influence the expression of Mas. It has been proposed that GPCRs in the RAS and KKS form functional homodimers and heterodimers162. A recent study by Wilson et al showed that biased AT1R receptor agonist [Sar1, Ile4, Ile8]-AngII negatively regulated cell surface B2R receptor by internalizing AT1R and B2R heterodimers180. B2R may regulate Mas receptor through a similar mechanism. Bioluminescence Resonance Eneergy

Transfer (BRET) technology can be used to detect the level of GPCR heterodimerization181.

Mechanisms for the beneficial effect of PGI2

Our studies have shown two aspects of the antithrombotic effect of PGI2: inhibiting platelet activation (in Bdkrb2-/- mice) and maintaining vascular homeostasis (in Klkb1-/- mice). However, it remained unclear at the molecular level by what mechanisms PGI2 produces these effects. Further study should address the following questions:

103

Mechanisms for reduced GPVI activation and integrin dependent spreading defect in Bdkrb2-/- mice.

We demonstrated that Bdkrb2-/- platelets had decreased ligation-dependent Syk phosphorylation upon convulxin stimulation. This observation can be recapitulated in WT platelets treated with PGI2 analog carbaprostacyclin (Figure

24A & 24B). The study by Loyau et al showed that elevation of cAMP by prostacyclin or forskolin inhibited platelet GPVI dimerization, which is an essential process for platelet GPVI activation136,138. These combined data suggests that the GPVI defect produced by PGI2 in Bdkrb2-/- platelets is upstream of Syk kinase in the pathway, and presumably during the GPVI dimerization process. This provides an explanation why the defect is specific to the GPVI pathway. We also showed the defective spreading in Bdkrb2-/- platelets is integrin-dependent. Further study should also determine which target of the signaling pathway is affected by PGI2 during integrin mediated outside-in shape change. A few novel PGI2/PKA dependent pathways have recently been identified: Aburima et al showed that phosphorylation of RhoA by PKA inhibits the association of RhoA with myosin phosphatase-targeting subunit 1 (MYPT1), therefore reducing the phosphorylation of myosin light chain (MLC) and platelet shape change182. Gegenbauer et al demonstrated that regulator of G-protein signaling 18 (RGS18) is a target for cAMP/PKA and its phosphorylation regulates

Gq dependent platelet activation99. These studies provide some potential candidates for our future study in Bdkrb2-/- platelets.

104

Molecular mechanisms by which PGI2 maintain vascular homeostasis.

In both Bdkrb2-/- and Klkb1-/- mice, we showed constitutively elevated PGI2 increase vascular mRNA of transcription factors Sirt1 and KLF4 (Figure 35). This observation is recapitulated in vitro in cultured mouse cardiac endothelial cells treated with carbaprostacyclin. Barbieri et al showed that COX-2 depleted mice are prothrombotic through reduced PGI2, decreased peroxisome proliferator- activated receptor δ (PPAR δ), inhibited Sirt1 and increased TF119. Our study showed that the reciprocal is operative in Klkb1-/- mice. However, it is novel for our study to show that PGI2 elevates KLF4. Two studies have shown that Sirt1 activation positively regulates KLF2 expression163,183. Further study should determine whether the effect of PGI2 on KLF4 expression is Sirt1 dependent or independent.

Conclusions

This study focuses on the mechanisms of thrombosis protection in two animal models derived from the KKS. It demonstrates how alterations in the KKS influence thrombosis risk through receptors of the RAS leading to renal elaboration of PGI2. It describes the antithrombotic effect of PGI2 on vascular and platelet functions, and emphasizes how PGI2 regulates vascular transcription factors. Understanding these pathways may lead to development of novel targets for antithrombotic therapy. Future investigations along this direction has promise for many sub-specialty of vascular and platelet research.

105

Literature Cited 1. Schmaier AH. The physiologic basis of assembly and activation of the plasma kallikrein/kinin system. Thromb Haemost. 2004;91:1-3. 2. Muller F, Mutch NJ, Schenk WA, et al. Platelet polyphosphates are proinflammatory and procoagulant mediators in vivo. Cell. 2009;139:1143-1156. 3. Kaplan AP, Austen KF. A prealbumin activator of prekallikrein. II. Derivation of activators of prekallikrein from active Hageman factor by digestion with plasmin. J Exp Med. 1971;133:696-712. 4. Shariat-Madar Z, Mahdi F, Schmaier AH. Identification and characterization of prolylcarboxypeptidase as an endothelial cell prekallikrein activator. J Biol Chem. 2002;277:17962-17969. 5. Kaplan AP, Joseph K, Shibayama Y, Reddigari S, Ghebrehiwet B, Silverberg M. The intrinsic coagulation/kinin-forming cascade: assembly in plasma and cell surfaces in inflammation. Adv Immunol. 1997;66:225-272. 6. Mori K, Nagasawa S. Studies on human high molecular weight (HMW) kininogen. II. Structural change of HMW kininogen by the action of human plasma kallikrein. J Biochem. 1981;89:1465-1473. 7. Leeb-Lundberg LM, Marceau F, Muller-Esterl W, Pettibone DJ, Zuraw BL. International union of pharmacology. XLV. Classification of the kinin receptor family: from molecular mechanisms to pathophysiological consequences. Pharmacol Rev. 2005;57:27-77. 8. Brown NJ, Gainer JV, Murphey LJ, Vaughan DE. Bradykinin stimulates tissue plasminogen activator release from human forearm vasculature through B(2) receptor-dependent, NO synthase-independent, and cyclooxygenase- independent pathway. Circulation. 2000;102:2190-2196. 9. D'Orleans-Juste P, de Nucci G, Vane JR. Kinins act on B1 or B2 receptors to release conjointly endothelium-derived relaxing factor and prostacyclin from bovine aortic endothelial cells. Br J Pharmacol. 1989;96:920-926. 10. Bjorkqvist J, Sala-Cunill A, Renne T. Hereditary angioedema: a bradykinin-mediated swelling disorder. Thromb Haemost. 2013;109. 11. Murphey LJ, Gainer JV, Vaughan DE, Brown NJ. Angiotensin-converting enzyme insertion/deletion polymorphism modulates the human in vivo metabolism of bradykinin. Circulation. 2000;102:829-832. 12. Hasan AA, Warnock M, Nieman M, et al. Mechanisms of Arg-Pro-Pro-Gly- Phe inhibition of thrombin. Am J Physiol Heart Circ Physiol. 2003;285:H183-193. 13. Harrison-Bernard LM. The renal renin-angiotensin system. Adv Physiol Educ. 2009;33:270-274.

106

14. Mehta PK, Griendling KK. Angiotensin II cell signaling: physiological and pathological effects in the cardiovascular system. Am J Physiol Cell Physiol. 2007;292:C82-97. 15. Nishimura H, Tsuji H, Masuda H, et al. Angiotensin II increases plasminogen activator inhibitor-1 and tissue factor mRNA expression without changing that of tissue type plasminogen activator or tissue factor pathway inhibitor in cultured rat aortic endothelial cells. Thromb Haemost. 1997;77:1189- 1195. 16. Vaughan DE, Lazos SA, Tong K. Angiotensin II regulates the expression of plasminogen activator inhibitor-1 in cultured endothelial cells. A potential link between the renin-angiotensin system and thrombosis. J Clin Invest. 1995;95:995-1001. 17. Abadir PM, Carey RM, Siragy HM. Angiotensin AT2 receptors directly stimulate renal nitric oxide in bradykinin B2-receptor-null mice. Hypertension. 2003;42:600-604. 18. Jones ES, Vinh A, McCarthy CA, Gaspari TA, Widdop RE. AT2 receptors: functional relevance in cardiovascular disease. Pharmacol Ther. 2008;120:292- 316. 19. Billet S, Aguilar F, Baudry C, Clauser E. Role of angiotensin II AT1 receptor activation in cardiovascular diseases. Kidney Int. 2008;74:1379-1384. 20. Paulis L, Steckelings UM, Unger T. Key advances in antihypertensive treatment. Nat Rev Cardiol. 2012;9:276-285. 21. Vickers C, Hales P, Kaushik V, et al. Hydrolysis of biological peptides by human angiotensin-converting enzyme-related carboxypeptidase. J Biol Chem. 2002;277:14838-14843. 22. Odya CE, Marinkovic DV, Hammon KJ, Stewart TA, Erdos EG. Purification and properties of prolylcarboxypeptidase (angiotensinase C) from human kidney. J Biol Chem. 1978;253:5927-5931. 23. Hilchey SD, Bell-Quilley CP. Association between the natriuretic action of angiotensin-(1-7) and selective stimulation of renal prostaglandin I2 release. Hypertension. 1995;25:1238-1244. 24. Sampaio WO, Souza dos Santos RA, Faria-Silva R, da Mata Machado LT, Schiffrin EL, Touyz RM. Angiotensin-(1-7) through receptor Mas mediates endothelial nitric oxide synthase activation via Akt-dependent pathways. Hypertension. 2007;49:185-192. 25. Hayashi N, Yamamoto K, Ohishi M, et al. The counterregulating role of ACE2 and ACE2-mediated angiotensin 1-7 signaling against angiotensin II stimulation in vascular cells. Hypertens Res. 2010;33:1182-1185.

107

26. Schmaier AH. The kallikrein-kinin and the renin-angiotensin systems have a multilayered interaction. Am J Physiol Regul Integr Comp Physiol. 2003;285:R1-13. 27. Souza Dos Santos RA, Passaglio KT, Pesquero JB, Bader M, Simoes ESAC. Interactions between angiotensin-(1-7), kinins, and angiotensin II in kidney and blood vessels. Hypertension. 2001;38:660-664. 28. Bomtempo CA, Santos GF, Santos RA, Campagnole-Santos MJ. Interaction of bradykinin and angiotensin-(1-7) in the central modulation of the baroreflex control of the heart rate. J Hypertens. 1998;16:1797-1804. 29. Greco AJ, Master RG, Fokin A, Jr., Baber SR, Kadowitz PJ. Angiotensin- (1-7) potentiates responses to bradykinin but does not change responses to angiotensin I. Can J Physiol Pharmacol. 2006;84:1163-1175. 30. Su JB, Barbe F, Crozatier B, Campbell DJ, Hittinger L. Increased bradykinin levels accompany the hemodynamic response to acute inhibition of angiotensin-converting enzyme in dogs with heart failure. J Cardiovasc Pharmacol. 1999;34:700-710. 31. Endlich K, Steinhausen M. Role of kinins and angiotensin II in the vasodilating action of angiotensin converting enzyme inhibition in rat renal vessels. J Hypertens. 1997;15:633-641. 32. Adams GN, LaRusch GA, Stavrou E, et al. Murine prolylcarboxypeptidase depletion induces vascular dysfunction with hypertension and faster arterial thrombosis. Blood. 2011;117:3929-3937. 33. AbdAlla S, Abdel-Baset A, Lother H, el Massiery A, Quitterer U. Mesangial AT1/B2 receptor heterodimers contribute to angiotensin II hyperresponsiveness in experimental hypertension. J Mol Neurosci. 2005;26:185-192. 34. AbdAlla S, Lother H, Quitterer U. AT1-receptor heterodimers show enhanced G-protein activation and altered receptor sequestration. Nature. 2000;407:94-98. 35. AbdAlla S, Lother H, el Massiery A, Quitterer U. Increased AT(1) receptor heterodimers in preeclampsia mediate enhanced angiotensin II responsiveness. Nat Med. 2001;7:1003-1009. 36. Abadir PM, Periasamy A, Carey RM, Siragy HM. Angiotensin II type 2 receptor-bradykinin B2 receptor functional heterodimerization. Hypertension. 2006;48:316-322. 37. Motta G, Rojkjaer R, Hasan AA, Cines DB, Schmaier AH. High molecular weight kininogen regulates prekallikrein assembly and activation on endothelial cells: a novel mechanism for contact activation. Blood. 1998;91:516-528.

108

38. Langhauser F, Gob E, Kraft P, et al. Kininogen deficiency protects from ischemic neurodegeneration in mice by reducing thrombosis, blood-brain barrier damage, and inflammation. Blood;120:4082-4092. 39. Merkulov S, Zhang WM, Komar AA, et al. Deletion of murine kininogen gene 1 (mKng1) causes loss of plasma kininogen and delays thrombosis. Blood. 2008;111:1274-1281. 40. Shariat-Madar Z, Mahdi F, Warnock M, et al. Bradykinin B2 receptor knockout mice are protected from thrombosis by increased nitric oxide and prostacyclin. Blood. 2006;108:192-199. 41. Dahlof B, Devereux RB, Kjeldsen SE, et al. Cardiovascular morbidity and mortality in the Losartan Intervention For Endpoint reduction in hypertension study (LIFE): a randomised trial against atenolol. Lancet. 2002;359:995-1003. 42. Yusuf S, Sleight P, Pogue J, Bosch J, Davies R, Dagenais G. Effects of an angiotensin-converting-enzyme inhibitor, ramipril, on cardiovascular events in high-risk patients. The Heart Outcomes Prevention Evaluation Study Investigators. N Engl J Med. 2000;342:145-153. 43. Olson JA, Jr., Shiverick KT, Ogilvie S, Buhi WC, Raizada MK. Angiotensin II induces secretion of plasminogen activator inhibitor 1 and a tissue metalloprotease inhibitor-related protein from rat brain astrocytes. Proc Natl Acad Sci U S A. 1991;88:1928-1932. 44. Senchenkova EY, Russell J, Almeida-Paula LD, Harding JW, Granger DN. Angiotensin II-mediated microvascular thrombosis. Hypertension;56:1089-1095. 45. Kaminska M, Mogielnicki A, Stankiewicz A, et al. Angiotensin II via AT1 receptor accelerates arterial thrombosis in renovascular hypertensive rats. J Physiol Pharmacol. 2005;56:571-585. 46. Ridker PM, Gaboury CL, Conlin PR, Seely EW, Williams GH, Vaughan DE. Stimulation of plasminogen activator inhibitor in vivo by infusion of angiotensin II. Evidence of a potential interaction between the renin-angiotensin system and fibrinolytic function. Circulation. 1993;87:1969-1973. 47. Fraga-Silva RA, Da Silva DG, Montecucco F, et al. The angiotensin- converting enzyme 2/angiotensin-(1-7)/Mas receptor axis: a potential target for treating thrombotic diseases. Thromb Haemost;108:1089-1096. 48. Radomski MW, Palmer RM, Moncada S. The anti-aggregating properties of vascular endothelium: interactions between prostacyclin and nitric oxide. Br J Pharmacol. 1987;92:639-646. 49. Kucharewicz I, Chabielska E, Pawlak D, Matys T, Rolkowski R, Buczko W. The antithrombotic effect of angiotensin-(1-7) closely resembles that of losartan. J Renin Angiotensin Aldosterone Syst. 2000;1:268-272.

109

50. Rajendran S, Chirkov YY, Campbell DJ, Horowitz JD. Angiotensin-(1-7) enhances anti-aggregatory effects of the nitric oxide donor sodium nitroprusside. J Cardiovasc Pharmacol. 2005;46:459-463. 51. Kucharewicz I, Pawlak R, Matys T, Pawlak D, Buczko W. Antithrombotic effect of captopril and losartan is mediated by angiotensin-(1-7). Hypertension. 2002;40:774-779. 52. Fraga-Silva RA, Pinheiro SV, Goncalves AC, Alenina N, Bader M, Santos RA. The antithrombotic effect of angiotensin-(1-7) involves mas-mediated NO release from platelets. Mol Med. 2008;14:28-35. 53. Fraga-Silva RA, Sorg BS, Wankhede M, et al. ACE2 activation promotes antithrombotic activity. Mol Med;16:210-215. 54. Fraga-Silva RA, Costa-Fraga FP, De Sousa FB, et al. An orally active formulation of angiotensin-(1-7) produces an antithrombotic effect. Clinics (Sao Paulo);66:837-841. 55. Schulz C, Massberg S. Platelets in atherosclerosis and thrombosis. Handb Exp Pharmacol:111-133. 56. Boilard E, Blanco P, Nigrovic PA. Platelets: active players in the pathogenesis of arthritis and SLE. Nat Rev Rheumatol;8:534-542. 57. Gay LJ, Felding-Habermann B. Contribution of platelets to tumour metastasis. Nat Rev Cancer;11:123-134. 58. Jain S, Harris J, Ware J. Platelets: linking hemostasis and cancer. Arterioscler Thromb Vasc Biol;30:2362-2367. 59. Bertozzi CC, Hess PR, Kahn ML. Platelets: covert regulators of lymphatic development. Arterioscler Thromb Vasc Biol;30:2368-2371. 60. Li Z, Delaney MK, O'Brien KA, Du X. Signaling during platelet adhesion and activation. Arterioscler Thromb Vasc Biol;30:2341-2349. 61. Varga-Szabo D, Pleines I, Nieswandt B. Cell adhesion mechanisms in platelets. Arterioscler Thromb Vasc Biol. 2008;28:403-412. 62. Vaidyula VR, Rao AK. Role of Galphaq and phospholipase C-beta2 in human platelets activation by thrombin receptors PAR1 and PAR4: studies in human platelets deficient in Galphaq and phospholipase C-beta2. Br J Haematol. 2003;121:491-496. 63. Varga-Szabo D, Braun A, Nieswandt B. Calcium signaling in platelets. J Thromb Haemost. 2009;7:1057-1066.

110

64. Crittenden JR, Bergmeier W, Zhang Y, et al. CalDAG-GEFI integrates signaling for platelet aggregation and thrombus formation. Nat Med. 2004;10:982-986. 65. Hardy AR, Jones ML, Mundell SJ, Poole AW. Reciprocal cross-talk between P2Y1 and P2Y12 receptors at the level of calcium signaling in human platelets. Blood. 2004;104:1745-1752. 66. Hirsch E, Bosco O, Tropel P, et al. Resistance to thromboembolism in PI3Kgamma-deficient mice. Faseb J. 2001;15:2019-2021. 67. Woulfe D, Jiang H, Mortensen R, Yang J, Brass LF. Activation of Rap1B by G(i) family members in platelets. J Biol Chem. 2002;277:23382-23390. 68. Chen H, Kahn ML. Reciprocal signaling by integrin and nonintegrin receptors during collagen activation of platelets. Mol Cell Biol. 2003;23:4764- 4777. 69. Watson SP, Auger JM, McCarty OJ, Pearce AC. GPVI and integrin alphaIIb beta3 signaling in platelets. J Thromb Haemost. 2005;3:1752-1762. 70. Schmaier AA, Zou Z, Kazlauskas A, et al. Molecular priming of Lyn by GPVI enables an immune receptor to adopt a hemostatic role. Proc Natl Acad Sci U S A. 2009;106:21167-21172. 71. Inoue O, Suzuki-Inoue K, Dean WL, Frampton J, Watson SP. Integrin alpha2beta1 mediates outside-in regulation of platelet spreading on collagen through activation of Src kinases and PLCgamma2. J Cell Biol. 2003;160:769- 780. 72. Li Z, Delaney MK, O'Brien KA, Du X. Signaling during platelet adhesion and activation. Arterioscler Thromb Vasc Biol. 2010;30:2341-2349. 73. Wang Z, Holly SP, Larson MK, et al. Rap1b is critical for glycoprotein VI- mediated but not ADP receptor-mediated alpha2beta1 activation. J Thromb Haemost. 2009;7:693-700. 74. Suzuki-Inoue K, Yatomi Y, Asazuma N, et al. Rac, a small guanosine triphosphate-binding protein, and p21-activated kinase are activated during platelet spreading on collagen-coated surfaces: roles of integrin alpha(2)beta(1). Blood. 2001;98:3708-3716. 75. Guidetti GF, Bernardi B, Consonni A, et al. Integrin alpha2beta1 induces phosphorylation-dependent and phosphorylation-independent activation of phospholipase Cgamma2 in platelets: role of Src kinase and Rac GTPase. J Thromb Haemost. 2009;7:1200-1206. 76. McCarty OJ, Larson MK, Auger JM, et al. Rac1 is essential for platelet lamellipodia formation and aggregate stability under flow. J Biol Chem. 2005;280:39474-39484.

111

77. Hartwig JH, Bokoch GM, Carpenter CL, et al. Thrombin receptor ligation and activated Rac uncap actin filament barbed ends through phosphoinositide synthesis in permeabilized human platelets. Cell. 1995;82:643-653. 78. Azim AC, Barkalow K, Chou J, Hartwig JH. Activation of the small GTPases, rac and cdc42, after ligation of the platelet PAR-1 receptor. Blood. 2000;95:959-964. 79. Schmidt VA, Scudder L, Devoe CE, Bernards A, Cupit LD, Bahou WF. IQGAP2 functions as a GTP-dependent effector protein in thrombin-induced platelet cytoskeletal reorganization. Blood. 2003;101:3021-3028. 80. Vidal C, Geny B, Melle J, Jandrot-Perrus M, Fontenay-Roupie M. Cdc42/Rac1-dependent activation of the p21-activated kinase (PAK) regulates human platelet lamellipodia spreading: implication of the cortical-actin binding protein cortactin. Blood. 2002;100:4462-4469. 81. Rozelle AL, Machesky LM, Yamamoto M, et al. Phosphatidylinositol 4,5- bisphosphate induces actin-based movement of raft-enriched vesicles through WASP-Arp2/3. Curr Biol. 2000;10:311-320. 82. Arthur WT, Petch LA, Burridge K. Integrin engagement suppresses RhoA activity via a c-Src-dependent mechanism. Curr Biol. 2000;10:719-722. 83. Flevaris P, Stojanovic A, Gong H, Chishti A, Welch E, Du X. A molecular switch that controls cell spreading and retraction. J Cell Biol. 2007;179:553-565. 84. Knight CG, Morton LF, Peachey AR, Tuckwell DS, Farndale RW, Barnes MJ. The collagen-binding A-domains of integrins alpha(1)beta(1) and alpha(2)beta(1) recognize the same specific amino acid sequence, GFOGER, in native (triple-helical) collagens. J Biol Chem. 2000;275:35-40. 85. Watanabe N, Bodin L, Pandey M, et al. Mechanisms and consequences of agonist-induced talin recruitment to platelet integrin alphaIIbbeta3. J Cell Biol. 2008;181:1211-1222. 86. Moser M, Nieswandt B, Ussar S, Pozgajova M, Fassler R. Kindlin-3 is essential for integrin activation and platelet aggregation. Nat Med. 2008;14:325- 330. 87. Petrich BG, Marchese P, Ruggeri ZM, et al. Talin is required for integrin- mediated platelet function in hemostasis and thrombosis. J Exp Med. 2007;204:3103-3111. 88. Arias-Salgado EG, Lizano S, Sarkar S, Brugge JS, Ginsberg MH, Shattil SJ. Src kinase activation by direct interaction with the integrin beta cytoplasmic domain. Proc Natl Acad Sci U S A. 2003;100:13298-13302.

112

89. Xi X, Flevaris P, Stojanovic A, et al. Tyrosine phosphorylation of the integrin beta 3 subunit regulates beta 3 cleavage by calpain. J Biol Chem. 2006;281:29426-29430. 90. Gong H, Shen B, Flevaris P, et al. G protein subunit Galpha13 binds to integrin alphaIIbbeta3 and mediates integrin "outside-in" signaling. Science. 2010;327:340-343. 91. Jin RC, Loscalzo J. Vascular Nitric Oxide: Formation and Function. J Blood Med;2010:147-162. 92. Chen LY, Mehta JL. Further evidence of the presence of constitutive and inducible nitric oxide synthase isoforms in human platelets. J Cardiovasc Pharmacol. 1996;27:154-158. 93. Denninger JW, Marletta MA. Guanylate cyclase and the .NO/cGMP signaling pathway. Biochim Biophys Acta. 1999;1411:334-350. 94. Jones CI, Barrett NE, Moraes LA, Gibbins JM, Jackson DE. Endogenous inhibitory mechanisms and the regulation of platelet function. Methods Mol Biol;788:341-366. 95. Marnett LJ, Rowlinson SW, Goodwin DC, Kalgutkar AS, Lanzo CA. Arachidonic acid oxygenation by COX-1 and COX-2. Mechanisms of catalysis and inhibition. J Biol Chem. 1999;274:22903-22906. 96. Gryglewski RJ. Prostacyclin among prostanoids. Pharmacol Rep. 2008;60:3-11. 97. Miggin SM, Kinsella BT. Investigation of the mechanisms of G protein: effector coupling by the human and mouse prostacyclin receptors. Identification of critical species-dependent differences. J Biol Chem. 2002;277:27053-27064. 98. Smolenski A. Novel roles of cAMP/cGMP-dependent signaling in platelets. J Thromb Haemost;10:167-176. 99. Gegenbauer K, Elia G, Blanco-Fernandez A, Smolenski A. Regulator of G- protein signaling 18 integrates activating and inhibitory signaling in platelets. Blood;119:3799-3807. 100. Geiger J, Nolte C, Butt E, Sage SO, Walter U. Role of cGMP and cGMP- dependent protein kinase in nitrovasodilator inhibition of agonist-evoked calcium elevation in human platelets. Proc Natl Acad Sci U S A. 1992;89:1031-1035. 101. Aszodi A, Pfeifer A, Ahmad M, et al. The vasodilator-stimulated phosphoprotein (VASP) is involved in cGMP- and cAMP-mediated inhibition of agonist-induced platelet aggregation, but is dispensable for smooth muscle function. Embo J. 1999;18:37-48.

113

102. Li Z, Ajdic J, Eigenthaler M, Du X. A predominant role for cAMP- dependent protein kinase in the cGMP-induced phosphorylation of vasodilator- stimulated phosphoprotein and platelet inhibition in humans. Blood. 2003;101:4423-4429. 103. Morrell CN, Matsushita K, Chiles K, et al. Regulation of platelet granule exocytosis by S-nitrosylation. Proc Natl Acad Sci U S A. 2005;102:3782-3787. 104. Irwin C, Roberts W, Naseem KM. Nitric oxide inhibits platelet adhesion to collagen through cGMP-dependent and independent mechanisms: the potential role for S-nitrosylation. Platelets. 2009;20:478-486. 105. Zhang G, Xiang B, Dong A, et al. Biphasic roles for soluble guanylyl cyclase (sGC) in platelet activation. Blood;118:3670-3679. 106. Loscalzo J. Nitric oxide insufficiency, platelet activation, and arterial thrombosis. Circ Res. 2001;88:756-762. 107. Yoshimoto H, Suehiro A, Kakishita E. Exogenous nitric oxide inhibits platelet activation in whole blood. J Cardiovasc Pharmacol. 1999;33:109-115. 108. Roberts W, Riba R, Homer-Vanniasinkam S, Farndale RW, Naseem KM. Nitric oxide specifically inhibits integrin-mediated platelet adhesion and spreading on collagen. J Thromb Haemost. 2008;6:2175-2185. 109. Schafer A, Wiesmann F, Neubauer S, Eigenthaler M, Bauersachs J, Channon KM. Rapid regulation of platelet activation in vivo by nitric oxide. Circulation. 2004;109:1819-1822. 110. Bunting S, Simmons PM, Moncada S. Inhibition of platelet activation by prostacyclin: possible consequences in coagulation and anticoagulation. Thromb Res. 1981;21:89-102. 111. Pedvis LG, Wong T, Frojmovic MM. Differential inhibition of the platelet activation sequence: shape change, micro- and macro-aggregation, by a stable prostacyclin analogue (Iloprost). Thromb Haemost. 1988;59:323-328. 112. Freedman JE, Sauter R, Battinelli EM, et al. Deficient platelet-derived nitric oxide and enhanced hemostasis in mice lacking the NOSIII gene. Circ Res. 1999;84:1416-1421. 113. Ozuyaman B, Godecke A, Kusters S, Kirchhoff E, Scharf RE, Schrader J. Endothelial nitric oxide synthase plays a minor role in inhibition of arterial thrombus formation. Thromb Haemost. 2005;93:1161-1167. 114. Jiang R, Wang S, Takahashi K, et al. Generation of a conditional allele for the mouse endothelial nitric oxide synthase gene. Genesis;50:685-692. 115. Cheng Y, Austin SC, Rocca B, et al. Role of prostacyclin in the cardiovascular response to thromboxane A2. Science. 2002;296:539-541.

114

116. Yu Y, Ricciotti E, Scalia R, et al. Vascular COX-2 modulates blood pressure and thrombosis in mice. Sci Transl Med;4:132ra154. 117. Seta F, Chung AD, Turner PV, Mewburn JD, Yu Y, Funk CD. Renal and cardiovascular characterization of COX-2 knockdown mice. Am J Physiol Regul Integr Comp Physiol. 2009;296:R1751-1760. 118. Riehl TE, He L, Zheng L, Greco S, Tollefsen DM, Stenson WF. COX-1(+/- )COX-2(-/-) genotype in mice is associated with shortened time to carotid artery occlusion through increased PAI-1. J Thromb Haemost;9:350-360. 119. Barbieri SS, Amadio P, Gianellini S, et al. Cyclooxygenase-2-derived prostacyclin regulates arterial thrombus formation by suppressing tissue factor in a sirtuin-1-dependent-manner. Circulation;126:1373-1384. 120. Ghosh M, Wang H, Ai Y, et al. COX-2 suppresses tissue factor expression via endocannabinoid-directed PPARdelta activation. J Exp Med. 2007;204:2053- 2061. 121. Major cardiovascular events in hypertensive patients randomized to doxazosin vs chlorthalidone: the antihypertensive and lipid-lowering treatment to prevent heart attack trial (ALLHAT). ALLHAT Collaborative Research Group. Jama. 2000;283:1967-1975. 122. Yang HY, Erdos EG. Second kininase in human blood plasma. Nature. 1967;215:1402-1403. 123. Hasan AA, Amenta S, Schmaier AH. Bradykinin and its metabolite, Arg- Pro-Pro-Gly-Phe, are selective inhibitors of alpha-thrombin-induced platelet activation. Circulation. 1996;94:517-528. 124. Tan Y, Keum JS, Wang B, McHenry MB, Lipsitz SR, Jaffa AA. Targeted deletion of B2-kinin receptors protects against the development of diabetic nephropathy. Am J Physiol Renal Physiol. 2007;293:F1026-1035. 125. Cervenka L, Harrison-Bernard LM, Dipp S, Primrose G, Imig JD, El-Dahr SS. Early onset salt-sensitive hypertension in bradykinin B(2) receptor null mice. Hypertension. 1999;34:176-180. 126. Hong SL. Effect of bradykinin and thrombin on prostacyclin synthesis in endothelial cells from calf and pig aorta and human umbilical cord vein. Thromb Res. 1980;18:787-795. 127. Santos RA, Simoes e Silva AC, Maric C, et al. Angiotensin-(1-7) is an endogenous ligand for the G protein-coupled receptor Mas. Proc Natl Acad Sci U S A. 2003;100:8258-8263. 128. Campbell DJ, Krum H, Esler MD. Losartan increases bradykinin levels in hypertensive humans. Circulation. 2005;111:315-320.

115

129. Seyedi N, Xu X, Nasjletti A, Hintze TH. Coronary kinin generation mediates nitric oxide release after angiotensin receptor stimulation. Hypertension. 1995;26:164-170. 130. Brosnihan KB, Li P, Ferrario CM. Angiotensin-(1-7) dilates canine coronary arteries through kinins and nitric oxide. Hypertension. 1996;27:523-528. 131. Luque M, Martin P, Martell N, Fernandez C, Brosnihan KB, Ferrario CM. Effects of captopril related to increased levels of prostacyclin and angiotensin-(1- 7) in essential hypertension. J Hypertens. 1996;14:799-805. 132. Grobe N, Elased KM, Cool DR, Morris M. Mass spectrometry for the molecular imaging of angiotensin metabolism in kidney. Am J Physiol Endocrinol Metab;302:E1016-1024. 133. Ashby B. Model of prostaglandin-regulated cyclic AMP metabolism in intact platelets: examination of time-dependent effects on adenylate cyclase and phosphodiesterase activities. Mol Pharmacol. 1989;36:866-873. 134. Coles B, Bloodsworth A, Eiserich JP, et al. Nitrolinoleate inhibits platelet activation by attenuating calcium mobilization and inducing phosphorylation of vasodilator-stimulated phosphoprotein through elevation of cAMP. J Biol Chem. 2002;277:5832-5840. 135. Zhang W, Colman RW. Thrombin regulates intracellular cyclic AMP concentration in human platelets through phosphorylation/activation of phosphodiesterase 3A. Blood. 2007;110:1475-1482. 136. Loyau S, Dumont B, Ollivier V, et al. Platelet glycoprotein VI dimerization, an active process inducing receptor competence, is an indicator of platelet reactivity. Arterioscler Thromb Vasc Biol;32:778-785. 137. Suzuki H, Murasaki K, Kodama K, Takayama H. Intracellular localization of glycoprotein VI in human platelets and its surface expression upon activation. Br J Haematol. 2003;121:904-912. 138. Jung SM, Moroi M, Soejima K, et al. Constitutive dimerization of glycoprotein VI (GPVI) in resting platelets is essential for binding to collagen and activation in flowing blood. J Biol Chem;287:30000-30013. 139. Komine N, Khang S, Wead LM, Blantz RC, Gabbai FB. Effect of combining an ACE inhibitor and an angiotensin II receptor blocker on plasma and kidney tissue angiotensin II levels. Am J Kidney Dis. 2002;39:159-164. 140. Ferrario CM, Jessup J, Gallagher PE, et al. Effects of renin-angiotensin system blockade on renal angiotensin-(1-7) forming enzymes and receptors. Kidney Int. 2005;68:2189-2196. 141. Best D, Senis YA, Jarvis GE, et al. GPVI levels in platelets: relationship to platelet function at high shear. Blood. 2003;102:2811-2818.

116

142. Bjorkqvist J, Jamsa A, Renne T. Plasma kallikrein: the bradykinin- producing enzyme. Thromb Haemost;110. 143. Shariat-Madar Z, Mahdi F, Schmaier AH. Recombinant prolylcarboxypeptidase activates plasma prekallikrein. Blood. 2004;103:4554- 4561. 144. Han ED, MacFarlane RC, Mulligan AN, Scafidi J, Davis AE, 3rd. Increased vascular permeability in C1 inhibitor-deficient mice mediated by the bradykinin type 2 receptor. J Clin Invest. 2002;109:1057-1063. 145. Abildgaard CF, Harrison J. Fletcher factor deficiency: family study and detection. Blood. 1974;43:641-644. 146. Renne T, Gruber A. Plasma kallikrein: novel functions for an old protease. Thromb Haemost;107:1012-1013. 147. Smith D, Gilbert M, Owen WG. Tissue plasminogen activator release in vivo in response to vasoactive agents. Blood. 1985;66:835-839. 148. Palmer RM, Ferrige AG, Moncada S. Nitric oxide release accounts for the biological activity of endothelium-derived relaxing factor. Nature. 1987;327:524- 526. 149. Revenko AS, Gao D, Crosby JR, et al. Selective depletion of plasma prekallikrein or coagulation factor XII inhibits thrombosis in mice without increased risk of bleeding. Blood;118:5302-5311. 150. Bird JE, Smith PL, Wang X, et al. Effects of plasma kallikrein deficiency on haemostasis and thrombosis in mice: murine ortholog of the Fletcher trait. Thromb Haemost;107:1141-1150. 151. Fang C, Stavrou E, Schmaier AA, et al. Angiotensin 1-7 and Mas decrease thrombosis in Bdkrb2-/- mice by increasing NO and prostacyclin to reduce platelet spreading and glycoprotein VI activation. Blood;121:3023-3032. 152. Renne T, Pozgajova M, Gruner S, et al. Defective thrombus formation in mice lacking coagulation factor XII. J Exp Med. 2005;202:271-281. 153. Iwaki T, Castellino FJ. Plasma levels of bradykinin are suppressed in factor XII-deficient mice. Thromb Haemost. 2006;95:1003-1010. 154. Breitenstein A, Stein S, Holy EW, et al. Sirt1 inhibition promotes in vivo arterial thrombosis and tissue factor expression in stimulated cells. Cardiovasc Res;89:464-472. 155. Hagedorn I, Schmidbauer S, Pleines I, et al. Factor XIIa inhibitor recombinant human albumin Infestin-4 abolishes occlusive arterial thrombus formation without affecting bleeding. Circulation;121:1510-1517.

117

156. Bhattacharjee G, Revenko AS, Crosby JR, et al. Inhibition of vascular permeability by antisense-mediated inhibition of plasma kallikrein and coagulation factor 12. Nucleic Acid Ther;23:175-187. 157. Bjorkqvist J, Sala-Cunill A, Renne T. Hereditary angioedema: a bradykinin-mediated swelling disorder. Thromb Haemost;109:368-374. 158. Gailani D, Renne T. The intrinsic pathway of coagulation: a target for treating thromboembolic disease? J Thromb Haemost. 2007;5:1106-1112. 159. Gailani D, Renne T. Intrinsic pathway of coagulation and arterial thrombosis. Arterioscler Thromb Vasc Biol. 2007;27:2507-2513. 160. Kleinschnitz C, Stoll G, Bendszus M, et al. Targeting coagulation factor XII provides protection from pathological thrombosis in cerebral ischemia without interfering with hemostasis. J Exp Med. 2006;203:513-518. 161. van der Meijden PE, Munnix IC, Auger JM, et al. Dual role of collagen in factor XII-dependent thrombus formation. Blood. 2009;114:881-890. 162. Lyngso C, Erikstrup N, Hansen JL. Functional interactions between 7TM receptors in the renin-angiotensin system--dimerization or crosstalk? Mol Cell Endocrinol. 2009;302:203-212. 163. Wu Z, Liu MC, Liang M, Fu J. Sirt1 protects against thrombomodulin down-regulation and lung coagulation after particulate matter exposure. Blood;119:2422-2429. 164. Borkowski JA, Ransom RW, Seabrook GR, et al. Targeted disruption of a B2 bradykinin receptor gene in mice eliminates bradykinin action in smooth muscle and neurons. J Biol Chem. 1995;270:13706-13710. 165. Nieman MT, Warnock M, Hasan AA, et al. The preparation and characterization of novel peptide antagonists to thrombin and factor VIIa and activation of protease-activated receptor 1. J Pharmacol Exp Ther. 2004;311:492-501. 166. Schmittgen TD, Zakrajsek BA, Mills AG, Gorn V, Singer MJ, Reed MW. Quantitative reverse transcription-polymerase chain reaction to study mRNA decay: comparison of endpoint and real-time methods. Anal Biochem. 2000;285:194-204. 167. Collister JP, Osborn JW. Area postrema lesion attenuates the long-term hypotensive effects of losartan in salt-replete rats. Am J Physiol. 1998;274:R357- 366. 168. Lakshmanan AP, Watanabe K, Thandavarayan RA, et al. Telmisartan attenuates oxidative stress and renal fibrosis in streptozotocin induced diabetic mice with the alteration of angiotensin-(1-7) mas receptor expression associated with its PPAR-gamma agonist action. Free Radic Res;45:575-584.

118

169. Joyner J, Neves L, Ferrario C, Brosnihan K. Administration of D-Alanine- [Ang-(1-7)] (A-779) Prior to Pregnancy in Sprague Dawley Rats Produces Antidiuresis in Late Gestation. J Am Soc Hypertens. 2008;2:425-430. 170. Elased KM, Cunha TS, Gurley SB, Coffman TM, Morris M. New mass spectrometric assay for angiotensin-converting enzyme 2 activity. Hypertension. 2006;47:1010-1017. 171. Zhou L, Schmaier AH. Platelet aggregation testing in platelet-rich plasma: description of procedures with the aim to develop standards in the field. Am J Clin Pathol. 2005;123:172-183. 172. Schmaier AH, Zuckerberg A, Silverman C, Kuchibhotla J, Tuszynski GP, Colman RW. High-molecular weight kininogen. A secreted platelet protein. J Clin Invest. 1983;71:1477-1489. 173. Liu J, Gao BB, Clermont AC, et al. Hyperglycemia-induced cerebral hematoma expansion is mediated by plasma kallikrein. Nat Med;17:206-210. 174. De La Cadena RA, Scott CF, Colman RW. Evaluation of a microassay for human plasma prekallikrein. J Lab Clin Med. 1987;109:601-607. 175. Hemker HC, Giesen P, Al Dieri R, et al. Calibrated automated thrombin generation measurement in clotting plasma. Pathophysiol Haemost Thromb. 2003;33:4-15. 176. Smith SA, Choi SH, Davis-Harrison R, et al. Polyphosphate exerts differential effects on blood clotting, depending on polymer size. Blood;116:4353- 4359. 177. Zhou X, Arthur G. Improved procedures for the determination of lipid phosphorus by malachite green. J Lipid Res. 1992;33:1233-1236. 178. Grobe N, Weir NM, Leiva O, et al. Identification of prolyl carboxypeptidase as an alternative enzyme for processing of renal angiotensin II using mass spectrometry. Am J Physiol Cell Physiol;304:C945-953. 179. Saifudeen Z, Dipp S, Fan H, El-Dahr SS. Combinatorial control of the bradykinin B2 receptor promoter by p53, CREB, KLF-4, and CBP: implications for terminal nephron differentiation. Am J Physiol Renal Physiol. 2005;288:F899-909. 180. Wilson PC, Lee MH, Appleton KM, et al. The arrestin-selective angiotensin AT1 receptor agonist [Sar1,Ile4,Ile8]-AngII negatively regulates bradykinin B2 receptor signaling via AT1-B2 receptor heterodimers. J Biol Chem;288:18872- 18884. 181. Kufareva I, Stephens B, Gilliland CT, et al. A novel approach to quantify G-protein-coupled receptor dimerization equilibrium using bioluminescence resonance energy transfer. Methods Mol Biol;1013:93-127.

119

182. Aburima A, Wraith KS, Raslan Z, Law R, Magwenzi S, Naseem KM. cAMP signaling regulates platelet myosin light chain (MLC) phosphorylation and shape change through targeting the RhoA-Rho kinase-MLC phosphatase signaling pathway. Blood;122:3533-3545. 183. Gracia-Sancho J, Villarreal G, Jr., Zhang Y, Garcia-Cardena G. Activation of SIRT1 by resveratrol induces KLF2 expression conferring an endothelial vasoprotective phenotype. Cardiovasc Res;85:514-519.

120