<%BANNER%>

Structure/Function Relationship in the Jak2 Kinase Domain

Permanent Link: http://ufdc.ufl.edu/UFE0021596/00001

Material Information

Title: Structure/Function Relationship in the Jak2 Kinase Domain
Physical Description: 1 online resource (98 p.)
Language: english
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

Subjects / Keywords: function, jak2, signaling, structure
Physiology and Pharmacology (IDP) -- Dissertations, Academic -- UF
Genre: Medical Sciences thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Janus Kinase 2 (Jak2) is one of four members of the Janus family of non-receptor tyrosine kinases. Jak2 plays a role in many physiologically relevant cellular signaling pathways and provides an essential link between the cell membrane and the nucleus. In the years since its discovery, Jak2 has been linked to a number of different disease states, including diabetes, atherosclerosis, cancer, and heart disease. These links underscore the need to fully understand Jak2 function. Like all proteins, the structure of Jak2 is intimately tied to its function. In this dissertation, we investigated two structural elements in the Jak2 kinase domain that significantly impact its function. Tyrosine 972 (Y972) is a phosphotyrosine within the Jak2 kinase domain. We demonstrated that the loss of Y972 phosphorylation significantly affected elements of Jak2 function like autophosphorylation and ligand dependent signaling processes. However, we also determined that the loss of Y972 phosphorylation did not irreversibly affect Jak2 function. We also investigated Serine 1120 in the Jak2 kinase domain. The mutation of this residue to Alanine significantly hindered several elements of Jak2 function, including autophosphorylation, growth hormone dependent signal transduction, and Jak2 dependent gene transcription. Through the investigation of these sites, we learn more about the critical connection between Jak2 structure and its function.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Thesis: Thesis (Ph.D.)--University of Florida, 2008.
Local: Adviser: Sayeski, Peter P.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2010-05-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2008
System ID: UFE0021596:00001

Permanent Link: http://ufdc.ufl.edu/UFE0021596/00001

Material Information

Title: Structure/Function Relationship in the Jak2 Kinase Domain
Physical Description: 1 online resource (98 p.)
Language: english
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

Subjects / Keywords: function, jak2, signaling, structure
Physiology and Pharmacology (IDP) -- Dissertations, Academic -- UF
Genre: Medical Sciences thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Janus Kinase 2 (Jak2) is one of four members of the Janus family of non-receptor tyrosine kinases. Jak2 plays a role in many physiologically relevant cellular signaling pathways and provides an essential link between the cell membrane and the nucleus. In the years since its discovery, Jak2 has been linked to a number of different disease states, including diabetes, atherosclerosis, cancer, and heart disease. These links underscore the need to fully understand Jak2 function. Like all proteins, the structure of Jak2 is intimately tied to its function. In this dissertation, we investigated two structural elements in the Jak2 kinase domain that significantly impact its function. Tyrosine 972 (Y972) is a phosphotyrosine within the Jak2 kinase domain. We demonstrated that the loss of Y972 phosphorylation significantly affected elements of Jak2 function like autophosphorylation and ligand dependent signaling processes. However, we also determined that the loss of Y972 phosphorylation did not irreversibly affect Jak2 function. We also investigated Serine 1120 in the Jak2 kinase domain. The mutation of this residue to Alanine significantly hindered several elements of Jak2 function, including autophosphorylation, growth hormone dependent signal transduction, and Jak2 dependent gene transcription. Through the investigation of these sites, we learn more about the critical connection between Jak2 structure and its function.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Thesis: Thesis (Ph.D.)--University of Florida, 2008.
Local: Adviser: Sayeski, Peter P.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2010-05-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2008
System ID: UFE0021596:00001


This item has the following downloads:


Full Text

PAGE 1

1 STRUCTURE/FUNCTION RELATIONSHIP IN THE JAK2 KINASE DOMAIN By ISSAM MCDOOM A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2008

PAGE 2

2 2008 Issam McDoom

PAGE 3

3 To my family; Mom, Dad, Sof, Anu and Ro. I love you guys.

PAGE 4

4 ACKNOWLEDGMENTS First and foremost, I thank my mentor, Dr. Pe ter P. Sayeski. His patience and guidance have formed the foundation of my graduate trai ning. Without these elements, I would not have succeeded. I also thank my supervisory committee. Dr s. Jeffrey Harrison, Hideko Kasahara, and David Ostrov have provided invaluable advice th at has enriched my graduate work. Their impartial perspectives have highlighted the st rengths and weaknesses of my work and have greatly facilitated my improvement. I thank all of the past and present members of the Sayeski laboratory for their technical guidance and friendship. I also thank Dr. Davi d Ostrov and Andrew Magis for their assistance with in silico modeling applications. Without their tec hnical assistance, this work would not be possible. I thank all of my friends for their support. Thank you for providing me with an outlet for my stress. My graduate training woul d not have been the same without it. Finally, I thank my wonderful family for be ing an inexhaustible source of love and support. Thank you for helping me through those days that I could not see the light at the end of the tunnel. There is no way I could ha ve completed this process without you.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS...............................................................................................................4 LIST OF TABLES................................................................................................................. ..........8 LIST OF FIGURES................................................................................................................ .........9 ABSTRACT....................................................................................................................... ............10 CHAPTER 1 INTRODUCTION..................................................................................................................11 Signal Transduction and the Jak/STAT Family......................................................................11 Janus Kinase 2................................................................................................................. .......12 Jak2 Structure................................................................................................................. .13 Jak Homology (JH) Domains..........................................................................................14 Non-tyrosine residues important for Jak2 function..................................................14 Tyrosine residues important for Jak2 function.........................................................15 Jak2-Dependent Signaling......................................................................................................17 Jak2 Activation-Ligand Depende nt vs. Ligand Independent..........................................17 Cell Surface Receptors....................................................................................................18 Cytokine receptor signaling.....................................................................................18 GPCR signaling........................................................................................................19 Adaptors and Regulatory Proteins...................................................................................20 SHP phosphatase family...........................................................................................20 SH2B family.............................................................................................................21 SOCS Family............................................................................................................22 Reactive oxygen species...........................................................................................22 Jak2 and ROS Pathology........................................................................................................23 Diabetes....................................................................................................................... ....23 Diabetic nephropathy...............................................................................................23 Diabetic cardiomyopathy.........................................................................................25 Atherosclerosis................................................................................................................26 Cardiac Ischemia-Reperfusion Injury..............................................................................29 2 METHODS........................................................................................................................ .....34 In Silico Molecular Modeling of Jak2....................................................................................34 Mass Spectrometry.............................................................................................................. ...34 Cell Lines..................................................................................................................... ...........34 Cell Culture................................................................................................................... ..........35 Site-Directed Mutagenesis......................................................................................................35 Transient Cell Transfections...................................................................................................35 Immunoprecipitation............................................................................................................ ...36

PAGE 6

6 Western Blotting............................................................................................................... ......36 Luciferase Assay............................................................................................................... ......37 Fluorescence Microscopy.......................................................................................................37 3 Y972 AND JAK2 ACTIVATION..........................................................................................39 Summary........................................................................................................................ .........39 Introduction................................................................................................................... ..........39 Results........................................................................................................................ .............42 Tyrosine 972 is a Site of Jak2 Autophosphorylation.......................................................42 Tyrosine 972 is Solvent Accessible.................................................................................42 Y972 Affects Jak2 Total Tyrosine Phos phorylation and Y1007 Phosphorylation.........43 The Loss of Y972 Phosphorylation Affects Jak2 Kinase Activity, But Does Not Confer a Dominant Negative Phenotype.....................................................................43 Tyrosine 972 Does Not Affect Li gand-Independent STAT1 Activation........................45 Tyrosine 972 Phosphorylation Does No t Affect Ligand-Independent Gene Expression....................................................................................................................45 Tyrosine 972 is Important for th e Jak2 Response to Angiotensin II...............................46 Tyrosine 972 has Differential Effects on Growth Hormone-Dependent Jak2 Total and Y1007 Phosphorylation.........................................................................................46 Phosphorylation at Y972 Does Not Affect SH2BMediated Y1007/Y1008 Phosphorylation...........................................................................................................47 The Loss of Y972 Phosphorylation Impair s Growth Hormone-Mediated STAT1 Nuclear Translocation..................................................................................................47 Discussion..................................................................................................................... ..........48 4 S1120 AND JAK2 FUNCTION.............................................................................................62 Summary........................................................................................................................ .........62 Introduction................................................................................................................... ..........62 Results........................................................................................................................ .............64 Serine 1120 is Solvent Accessible...................................................................................64 Serine 1120 is Critical fo r Jak2 Autophosphorylation-...................................................65 The S1120A Mutant Displays a Mild Inhibitory Phenotype...........................................65 The Jak2 S1120A Mutation Hinders Ligand -Independent Gene Transcription-............66 Growth Hormone-Dependent Jak2 Activation is Dependent on S1120..........................66 Serine 1120 is Critical for SH2BMediated Jak2 Activation-......................................67 The S1120A Mutation Abolishes Growth Ho rmone-Mediated Gene Transcription......68 Discussion..................................................................................................................... ..........68 5 DISCUSSION..................................................................................................................... ....79 Overview....................................................................................................................... ..........79 Tyrosine 972 and Jak2 Function.............................................................................................79 Y972 and ROS Pathology.......................................................................................................80 Cardiac Ischemia Reperfusion Injury..............................................................................80

PAGE 7

7 Diabetic Nephropathy......................................................................................................82 Atherosclerosis................................................................................................................83 S1120 and Jak2 Function........................................................................................................83 S1120 and Growth Hormone-Dependent Pathophysiology...................................................84 Y972 vs. S1120................................................................................................................. ......86 Reflections.................................................................................................................... ..........87 Y972........................................................................................................................... .....88 S1120.......................................................................................................................... .....88 LIST OF REFERENCES............................................................................................................. ..92 BIOGRAPHICAL SKETCH.........................................................................................................98

PAGE 8

8 LIST OF TABLES Table page 5-1 A comparison of S1120A-dependent a nd Y972F-dependent effects on several categories of Jak2 function................................................................................................90

PAGE 9

9 LIST OF FIGURES Figure page 1-1 Jak Homology (JH) domains.............................................................................................33 3-1 Tyrosine 972 is a site of Jak2 autophosphorylation...........................................................52 3-2 Tyrosine 972 is solvent accessible.....................................................................................53 3-3 The effect of Y972 phosphorylation on Jak2 total and Y1007 phosphorylation...............54 3-4 Tyrosine 972 phosphorylation affects Ja k2 kinase function, but not its substrate properties or Jak2 dimerization..........................................................................................55 3-5 The loss of tyrosine 972 phosphorylati on does not affect Jak2-mediated STAT1 activation. ................................................................................................................... .......56 3-6 Tyrosine 972 does not affect lig and-independent gene expression...................................57 3-7 Tyrosine 972 is critical for angiot ensin II-dependent Jak2 phosphorylation.....................58 3-8 Tyrosine 972 differentially affect s growth hormone-dependent Jak2 total phosphorylation and Y1007 phosphorylation....................................................................59 3-9 Jak2 can be activated by SH2Bin the absence of tyrosine 972 phosphorylation..........60 3-10 Tyrosine 972 phosphorylation affect s GFP-STAT1 nuclear translocation........................61 4-1 Serine 1120 is solvent accessible.......................................................................................72 4-2 The importance of S1120 for Jak2 autophosphorylation...................................................73 4-3 The Jak2 S1120A mutant displays a mild inhibitory phenotype.......................................74 4-4 Serine 1120 is important for liga nd-independent gene transcription.................................75 4-5 Growth hormone-dependent Jak2 activati on is eliminated by the S1120A mutation.......76 4-6 Serine 1120 is critical for SH 2B-beta mediated Jak2 activation.......................................77 4-7 Serine 1120 is important for growth hor mone dependent gene transcription. .................78 5-1 The relative locations of Y972 and S1120 in murine Jak2................................................91

PAGE 10

10 Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy STRUCTURE/FUNCTION RELATIONSHIP IN THE JAK2 KINASE DOMAIN By Issam McDoom May, 2008 Chair: Peter P. Sayeski Major: Medical Sciences-Physiology and Pharmacology Janus Kinase 2 (Jak2) is one of four members of the Janus family of non-receptor tyrosine kinases. Jak2 plays a role in many physiol ogically relevant cellula r signaling pathways and provides an essential link between the cell membrane and the nucle us. In the years since its discovery, Jak2 has been linked to a number of different dis ease states, including diabetes, atherosclerosis, cancer, and heart disease. Th ese links underscore the n eed to fully understand Jak2 function. Like all proteins, the structure of Jak2 is intimatel y tied to its function. In this dissertation, we investigated two st ructural elements in the Jak2 ki nase domain that significantly impact its function. Tyrosine 972 (Y972) is a phosphotyrosine within th e Jak2 kinase domain. We demonstrated that the loss of Y972 phosphorylation significantly affected elements of Jak2 function like autophosphorylation an d ligand dependent signaling processes. However, we also determined that the loss of Y972 phosphorylation did not irreversibly affect Jak2 function. We also investigated Serine 1120 in the Jak2 kinase domain. The muta tion of this residue to Alanine significantly hindered several el ements of Jak2 function, includi ng autophosphorylation, growth hormone dependent signal transduction, and Jak2 dependent gene transcription. Through the investigation of these sites, we learn more about the critical connection between Jak2 structure and its function.

PAGE 11

11 CHAPTER 1 INTRODUCTION1 Signal Transduction and the Jak/STAT Family In multicellular organisms, the ability of cells to communicate with one another is absolutely essential for the coordination of physio logical processes. Most of the major human health issues in our society today have their f oundations in aberrant cell signaling. For instance, diabetes mellitus is the result of ei ther a lack of sensitivity to the biological message of insulin or the failure to produce this message in the first place. All forms of can cer, regardless of their tissue of origin, are commonly defined by dysfunc tional cellular growth signaling. Finally, the progression of cardiovascular disease is inextricably linked to pathological signaling events in the vascular wall and the myocardium. These examples highlight the fact that a profound understanding of cellular signaling pathways is abso lutely critical to our ability to combat the major human health issues of our time. Discovered in the early 1990s, the Janus family of tyrosine kinases is an indispensable component of cellular communication. This family consists of four members; Jak1, Jak2, Jak3, and Tyk2. Jak1, Jak2 and Tyk2 are ubiquitously e xpressed while Jak3 is mainly expressed in hematopoietic cells. Through a series of protei n phosphorylation events, th ese kinases relay the message represented by receptor/hormone binding at the cell surface to the cell nucleus. The ultimate outcome of Janus kinase dependent signal tran sduction is alteration of gene transcription patterns in the nucleus. This disse rtation will focus on Janus Kinase 2 (Jak2) and the structure-function relationshi p that governs this proteins ro le in cellular communication. Specifically, we will discuss two investigations into Jak2 structural elements. We will then discuss then discuss the relevance of these fi ndings to several Jak2 de pendent disorders. 1 A portion of Jak2 and Reactive Oxygen Species: A Complex Relationship is reprinted with permission from Bentham Science Publishers Ltd.

PAGE 12

12 Janus Kinase 2 In 1991, shortly after the discovery of the first member of the Janus Kinase family, the Jak2 gene was cloned (1). It wa s identified as a member of the Janus family by virtue of its possession of both a kinase and pseudokinase domai n, a trademark feature of the Jak family (1). Initially, Jak2 was characterized as a mediator of cytokine and grow th factor signaling (2). In 1993, erythropoietin (Epo) was identified as the fi rst ligand to activate Ja k2 by binding to the Epo receptor (3). Subsequent studies implicat ed Jak2 in the signaling pathways of growth hormone, interleukin 3, interferon gamma, prolac tin, and the GP130 receptor (4-8). Thus, the early history of Jak2 only portrays it as a mediator of cytokine signaling. As time passed, accumulating evidence subverted this narrow view of Jak2 and revealed that it is a much more versatile signaling mediat or than previously im agined. In 1994, it was shown that Jak2 can be activated in respons e to thrombin, a ligand that binds a seventransmembrane receptor. This report was a signif icant development in the Jak2 field as it was a departure from the original cytokine model of Jak2 activation. It linke d Jak2 activation to a whole new set of cellular processes, li ke G-protein coupled receptor activation, diacylglycerol/PKC signaling, and ce llular changes in IP3/Ca+2 (9). In 1995, this trend was continued by discovery that an giotensin II, acting through the AT1 receptor, could activate Jak2 (10). The relevance of Jak2 to human health grew steadily with every new discovery. Each new pathway in which Jak2 was implicated bolstered its image as a therapeutic target. For instance, the connection of Jak2 to both thrombin and angiot ensin II signaling makes it an attractive target for treating atherosclerosis, a disease in which bo th of these pathways have been shown to play roles. One of the most monumental developm ents in the history of the Jak2 field was the discovery in 1998 that Jak2 coul d be activated by r eactive oxygen species (11). This finding

PAGE 13

13 situated Jak2 within the field of oxidative stress signaling, an eve r-growing subfield within the signal transduction literatur e. It also exponentially increase d its pathological relevance as so many prominent human health issues have their foundations in oxidative stress. Another major event in the Jak2 field came in 2005, with the discovery of the Jak2 V617F point mutation and its relevance to myeloproliferative disorders (12) This finding cemented the position of Jak2 as a molecule that is critical to human health in many different ways. By elucidating the links between Jak2 and human health, the history of Jak2 readily highlights the need for the continuing study of this protein. In order to further our understanding of the role of Jak2 in human pathology, we must first understand the intricacies of Jak2 function on the molecular and cellular levels Therefore, the next two sect ions will be discussions of Jak2 structure and signaling. The rela tionship between structure and function is the dominant force that governs all Jak2 interactions. Upon this in trinsic foundation, the extrinsic elements that compose Jak2-dependent signaling combine to link the basic biochemical process of phosphorylation to the larger real m of cellular physiology and ultimately, human health as a whole. Jak2 Structure Like all proteins, the structur e of Jak2 dictates its functi on. In the years since Jak2 was discovered, Jak2 structure has been st udied intensely in an effort to identify elements that have profound functional implications for this protein. This endeavor is necessary if we are to fully understand the role of Jak2 in human physiol ogy and pathophysiology. Thus, the following sections are devoted to a disc ussion of Jak2 structure and its intimate relationship with Jak2 function. Specifically, we will discuss 1.) the or ganization of Jak2 primary structure into Jak Homology domains, 2.) functionally important amino acid residues (nontyrosine), and 3.) phosphotyrosine elements of Jak2 functional regulation.

PAGE 14

14 Jak Homology (JH) Domains The primary structures of all Jak family members share seven regions of structural similarity termed Jak Homology (J H) domains (Fig.1). The JH do mains are definitive structural features of the Jak family. The JH1 domain is located on the c-terminal end of the molecule and possesses the kinase activity of th e protein. Immediately n-terminal to the JH1 domain is the JH2, or pseudokinase domain. This domain is an inverse replicate of the kinase domain. Its juxtaposition to the JH1 do main is the source of Janus family name as this arrangement is reminiscent of Janus, the two-faced Roman god. The pseudokinase domain lacks kinase function, but it is thought to inhi bit the kinase function of the JH1 domain (13). The JH3-JH4 domains encode a putative SH2 domain. Fina lly, the JH6-JH7 domains compose the FERM domain, which has been shown to be necessary fo r Jak/receptor interactions. Within these seven regions, several amino acid residues have been identified as being important for regulating proper Jak2 function. These amino acids can be broken up into two classes; non-tyrosine residues and phosphotyrosines. Non-tyrosine residues important for Jak2 function Several amino acid residues within the ki nase and pseudokinase domain have been identified as being critical for maintaining pr oper Jak2 function. In the Jak2 kinase domain, E1024 and R1113 are thought to form a hydrogen bond th at is important for the ability of Jak2 to mediate angiotensin II signaling (14). These re sidues were also shown to be important for ligand-independent Jak2 autophosphorylation (14) Glutamic acid 1024 is also thought to interact with W1020 and di sruption of this interaction has be en shown to produce an inhibitory phenotype in Jak2 (14). Similarly, the intera ction between W1038 and E1046 has also been shown to be important for preventing an inhibito ry Jak2 phenotype (15). Finally, E1046 by itself

PAGE 15

15 has been shown to be important for Jak2 kinase function (16). The e ffects of mutating these amino acids underscore the delicate nature of the st ructural integrity of th e Jak2 kinase domain. In addition to these residues, structural elements within the pseudokinase domain also play a critical role in maintaining proper Jak2 function. Whereas disrupt ion of the kinase domain has been shown to produce a loss of function, structur al anomalies in the pseudokinase domain tend to have the opposite effect of enhancing Ja k2 function. A possible explanation for this phenomenon is that the pseudokinase domain is normally thought to inhibit Jak2 kinase function (17). Thus, structural abnormalities in this dom ain may relieve the inhibition. The Jak2 V617F point mutation is the most notable structural abnormality in the pseudokinase domain. Discovered in 2005 in patients with myeloproliferative disorders, this mutation is the result of an acquired G to T transversion in the Jak2 allele. The mutation leads to constitutive Jak2 phosphorylation and causes cellular hypersensitivity to cytokine stimulation, especially by erythropoietin (17). The V617F mutation was also linked to incr eases in gene expression of 13 genetic markers of polycythemia vera, a myelopro liferative disorder (18) In fact the V617F mutation is itself considered a marker for polyc ythemia vera, as nearly all patients with this disease have the V617F mutation (17). Thus the Jak2 V617F mutati on and its physiological consequences underscores the importance of the Jak2 structure/function relationship. Tyrosine residues import ant for Jak2 function Autophosphorylation is a central biochemical process in Jak2 signaling. This event transcends differences in cell t ype, receptor type, and activating stimulus as one of the common elements that binds all Jak2-dependent signali ng pathways together. There are 49 tyrosine residues in Jak2 and currently, at least eight of these residues are known to be phosphorylated (Y119, Y201, Y221, Y570, Y813, Y966, Y1007, Y1008). Mo st of these sites play important roles in regulating Jak2 function. The first sites to be discove red were Y1007 and Y1008 (19).

PAGE 16

16 These two tyrosines are phosphorylated in ta ndem, but only Y1007 has been shown to be important for Jak2 function. Tyrosine 1007 lies in the Jak2 activation loop, a substructure that blocks the ATP binding site of the kinase dom ain when Jak2 is inactive. When Y1007 is phosphorylated, the activation loop shifts and Jak2 becomes capable of transferring the phosphate from ATP to one of its substrates Therefore, it is thought that Y1007 phosphorylation is necessary for maximal Jak2 activation (19). Phosphotyrosines also regulate Jak2 func tion by serving as binding sites for Jak2 regulatory proteins. For instan ce, members of the Suppressors of Cytokine Signaling (SOCS) protein family bind to phosphorylated Y1007 in or der to terminate Jak2 signaling (20). Tyrosine 813 is another example of a protein binding site. In this case, phosphorylated Y813 serves as a binding site for SH2B, a potent Jak2 activator (21). This protein is capable of binding to Jak2 at other sites. However, these interactions produce an inhibito ry effect with respect to Jak2 activation. It is th e binding event at Y 813 that produces SH2Bdependent Jak2 activation. Tyrosines 221 and 570 have been shown to modulate Jak2 kinase f unction (22). These residues were shown to be phosphorylated in re sponse to growth hormone treatment (22). Phosphorylation of Y221 was shown to in crease Jak2 kinase activity while Y570 phosphorylation was shown to decrease kinase act ivity (22). Interestingly, Y570 is located within the Jak2 pseudokinase domain and the fact that its phosphorylation decreases Jak2 kinase activity strengthens the argument that the pseud okinase domain is respon sible for inhibiting the Jak2 kinase domain. Tyrosines 201 and 119 are located within the Jak2 FERM domai n. As such, they are both associated with receptor/Jak2 interactions. Ou r laboratory has shown that the phosphorylation of Y201 is a necessary precursor for Jak2 interaction with the AT1 receptor (23). This tyrosine

PAGE 17

17 serves as a binding site for SHP-2, a tyrosine phos phatase (23). This protein then serves as an adaptor for Jak2 interaction with the AT1 receptor (23). In contrast, Y119 plays a negative role in Jak2/receptor interactions. Specifically, the phosphorylation of Y119 has been shown to cause the dissociation of Jak2 from the erythropoietin receptor (24). In summary, phosphotyrosines play dive rse roles in regulating Jak2 function. Phosphorylation events can have both positive and negative consequences with regard to the propagation of a Jak2-dependent si gnal. Thus far, we have ma inly discussed the intrinsic elements that play important roles in regulating Jak2 function. However, Jak2 does not exist in a vacuum and several external factors influence Jak2 function as profoundly as the structural elements previously discussed. Therefore, in the next section, we will discuss the external factors that compose a Jak2dependent signaling pathway. Jak2-Dependent Signaling The primary cellular role of Jak2 is to relay signals from the cell surface to the nucleus. The versatility of Jak2-dependent signaling arises from the fact that many different proteins participate in this signaling pro cess, allowing for a wide range of combinations to produce a unique signal. Included in this list of protei ns are cell-surface receptors, adaptor/regulatory proteins, and members of the Signal Transducers a nd Activators of Transcription (STAT) family. These proteins are all critical compon ents of Jak2-dependent signaling. Jak2 Activation-Ligand Depende nt vs. Ligand Independent Autophosphorylation is a central component of Jak2 dependent si gnaling. It is important to remember that, in addition to being a kinase, a Jak2 protein is also a substrate for other Jak2 proteins. Under normal physiological conditions, Ja k2 activation must be achieved with the aid of an external signaling component, such as a cell surface receptor. However, when Jak2 is expressed at levels that exceed those seen in normal physiology, it is capable of activating itself

PAGE 18

18 in the absence of external stim ulation. This process is termed ligand independent activation and it is primarily a biochemical phenomenon. While ligand independent ac tivation does not have the physiological relevance of ligand dependent activa tion, it is a useful trai t that is exploited to study the intrinsic biochemical function of Jak2. In this dissertation, we use both ligand dependent and ligand independent methods of activating Jak2. Ou r ligand independent experiments were done to evaluate the manner in which structural changes affect the inherent activity of Jak2. Ligand dependent experiments we re then performed to extrapolate these effects to larger biochemical systems, represen ted by Jak2 dependent signaling pathways. Cell Surface Receptors Jak2-dependent signaling begins with ligand/re ceptor interactions at the cell surface. Hormones like angiotensin II and growth hormone represent information that is being relayed from one cell to another. When these hormones bind to their cell-surface receptors, the receptors activate Jak2-dependent signaling pathways, ultimately transforming this extracellular information into cellular changes in gene transcri ption. Jak2 is activated by three different types of cell-surface receptors; cytokine receptors, tyrosine kinase gr owth factor receptors, and Gprotein coupled receptors (GPCRs). Cytokine receptor signaling The most well-understood model of Jak2-dependent signaling is the cytokine model. In this model, Jak2 is constitutively bound to th e cytokine receptor through receptor/FERM domain interactions (25,26). Cytokine receptors, like the growth hormone receptor are thought to exist as dimers that undergo conformati onal changes as a result of ligan d/receptor association (27). These changes include rotations of the cytokine receptor subunits that are thought to put the constitutively associated Jak2 molecules in close proximity to one another (27). This process is thought to facilitate Jak2 aut ophosphorylation (27). Jak2 then phosphorylates the cytoplasmic

PAGE 19

19 tail of the cytokine receptor, allowing for the recr uitment of STAT proteins to the receptor (27). The recruited STAT proteins are phosphorylat ed by Jak2. The phosphorylated STAT proteins dimerize through reciprocal SH2 domain interactions These STAT dimers then translocate to the nucleus, where they bind promoter elements and alter gene transcription patterns. GPCR signaling The GPCR model of Jak2-depende nt signaling is a much more recent development than the cytokine model. Unlike cytokine receptors, GPCRs are not composed of dimerized subunits. Rather, these receptors consist of a single polype ptide chain that spans the plasma membrane seven times and have both intracellular and extrace llular components (28). The intracellular tail of the receptor is coupled to a hetorotrimeric G protein (28). These gua nosine nucleotide binding proteins are composed of 3 subunits: and As in the cytokine model, the initial ev ent in the GPCR model is a ligand/receptor association at the cell surface. Through a series of conformati onal changes, this event is translated into the GPCR-dependent activation of the coupled G-protein (28). The G protein then divides into and subunits. These subunits can then activate cytoplasmic Jak2 molecules (29,30). Upon activati on, the cytoplasmic Jak2 proteins bind to the cytoplasmic portion of the receptor tail via their FERM domains. Thus, this model differs from the cytokine model in that the Jak2 molecules are not constitu tively bound to the intrac ellular portion of the receptor. Another key difference between the tw o models is that, in the GPCR model, STAT proteins are thought to be re cruited to phosphotyrosines on Ja k2 as opposed to the cytokine model, in which they are recruite d to sites on the cytoplasmic tail of the cytokine receptor. Once recruited, STAT proteins are phos phorylated by Jak2 and the subse quent events are the same as those in the cytokine model.

PAGE 20

20 While the GPCR model is not as well-char acterized as the cytokine model of Jak2dependent signaling, the evidence that has accu mulated has shown this model to be quite versatile. In contrast to the cytokine mode l, which outlines one general mechanism for Jak2 activation by this class of receptors, the GP CR model incorporates multiple modes of Jak2 activation. In rat aortic smooth muscle cells, it has been shown th at Jak2 can be activated by the G subunit of heterotrimeric G proteins coupled to the AT1 receptor (31). Additionally, there is evidence of Jak2 activation by the G q subunit of G proteins (29). Fi nally, there is evidence that GPCRs, like the AT1 receptor, can activate Jak2 through a calcium-dependent signaling pathway involving the tyrosine kinase PYK2 (31). These different modes of activation add a great deal of diversity to the GPCR model. As more GPCRs ar e found to activate Jak2, it will be interesting to see the mode of activ ation used by each. Adaptors and Regulatory Proteins While the backbone of Jak2-dependent signa ling is composed of receptors, Jak2 and STAT proteins, several other protein families play modulatory roles in Ja k2-dependent signaling. The three main protein families are the SHP pho sphatase family, the SH2B family, and the SOCS family. Members of each of thes e families combine to modify the basic receptor/Jak2/STAT signaling axis. SHP phosphatase family The SHP phosphatase family consists of two members: SHP-1 and SHP-2. These tyrosine phosphatases contain two SH2 domains and a ph osphatase domain. Th is protein family represents a versatile component of Jak/STAT signaling as it has been implicated in both facilitative and negative regulat ory mechanisms in Jak2-dependent pathways. For instance, while SHP-1 has been shown to act as a classica l phosphatase in the context of angiotensin II signaling, SHP-2 has also been shown to assist angiotensin II-dependent Jak2 activation (23).

PAGE 21

21 Specifically, our lab has shown th at SHP-2 binds to Y201 in the Jak2 JH7 domain and facilitates Jak2/AT1 receptor co-association (23). Howeve r, SHP-2 can also function as a classical phosphatase and has been shown to dephosphorylate Y 1007 in Jak2 (32). In the case of prolactin signaling, this event leads to decreased Jak2 inhibition via SOCS-1 and thus enhancement of the prolactin signal (32). SHP-2 can also play a negative regulatory role and has been shown to dephosphorylate both the growth hormone receptor and STAT5, leading to the termination of the Jak/STAT signal (33). In he matopoietic stem cells, SHP-2 dependent STAT5 dephosphorylation is important for regulating cell survival (33). Thus, these phos phatases play diverse regulatory roles with regard to Jak2-dependent signaling SH2B family The SH2B protein family consists of three members: APS, Lnk, and SH2B (34). Each family member possesses an SH2 domain, a pleckstrin homology domain, and a dimerization domain (34). There are seve ral different isoforms of th e SH2B protein, and the SH2Bisoform is known to bind Jak2 at phosphor ylated Y813 (21). SH2Bis a potent Jak2 activator and has been shown to enhance growth hormone-dependent Jak2 activation (21). It is thought that SH2Bpromotes Jak2 activation by facilitating Jak2 dimer form ation (35). Interestingly, SH2Bplays a dual role as it can both inhibit and promote Jak2 activation. Jak2 activation is achieved through the SH2 domain-dep endent interaction with Y813 while inhibition is thought to occur through a non-SH2 domain-dep endent interaction with Jak2 (34). The current model of Jak2/SH2Binteraction suggests that non-Y813 de pendent interactions keep SH2Bin close proximity to Jak2 in the ev ent that Y813 becomes phosphorylated (34). Once this phosphorylation event occurs, the SH2 domain of SH2Bbinds phosphorylated Y813 (34). SH2Bthen enhances Jak2 activation.

PAGE 22

22 SOCS Family Members of the Suppressors of Cytokine Signaling (SOCS) family are inducible repressors of Jak/STAT signaling. As their name implies, these proteins were first characterized as inhibitors of Jak2-dependent signaling in response to cytokine activation (36). However, it has also been shown that SOCS-3 expression is inducible by angiotensin II, suggesting that these proteins may also inhibit GPCRdependent Jak/STAT signaling ( 37). SOCS protein expression is induced by the propagation of a Jak/STAT signa l (36). SOCS proteins then bind to Jak2 at phosphorylated Y1007 and promote ubiquitin-mediated degradation (20). These proteins also inhibit Jak/STAT signaling at the level of STAT proteins. It has been shown that SOCS-7 can inhibit both STAT3 and ST AT5 phosphorylation (38). Reactive oxygen species As previously mentioned, it was discovered in 1998 that Jak2 is activated in response to the presence of reactive oxygen species (ROS) (11). It is th ought that ROS indirectly activate kinases by reversibly inactivati ng phosphatases, which then lead s to kinase activation (39). Cellular sources of ROS include membrane bound enzymes like NADPH oxidase, the mitochondria, and oxidized lipids. The field of ROS signaling is rapidly growing and the traditional view of these molecules as mere byproducts of cellular respiration is becoming increasingly archaic. It is clear that these species play a critical role in relaying cellular information and have even been compared to traditional second-messenger molecules. Therefore, the next sections will discuss the role of ROS-dependent Jak2 signaling in several prevalent human diseases, as outlined in the review, Jak2 and Reactive Oxygen Species: A Complex Relationship (40)

PAGE 23

23 Jak2 and ROS Pathology Diabetes Diabetes mellitus is a complex set of disord ers that are intertwined by the common thread of poor glucose metabolism. At the heart of this condition is the peptide hormone insulin. Type I diabetes is characterized by an in ability to secrete insulin from pancreatic beta cells. In the case of Type II diabetes, it is the abil ity of cells to respond to insulin that has been compromised. In both cases, the lack of insulin action results in abnormally high blood glucose levels. Ultimately, all diabetic complications can tr ace their lineage back to this hyperg lycemic state. These effects include, but are not limited to, macular degenerati on, renal disease, athero sclerosis, and heart disease. Therefore, it is imperative that diab etic patients properly co ntrol their blood glucose levels. While this is obviously the first and mo st important line of defense against diabetic complications, understanding the mechanisms behind diabetic pathologies w ill greatly facilitate the development of therapeutics. The interaction between ROS and Jak2 is a significant component of the mechanisms behind a number of these complications. In this section, we will discuss the implications of the Jak2/ROS rela tionship for the development of diabetic nephropathy and diabetic cardiomyopathy. Diabetic nephropathy Diabetic nephropathy is a diso rder of the renal glomerulus that is characterized by the development of glomerular lesions (41). This disease is the most common precursor to End Stage Renal Disease (41). The early phase of diabetic nephropathy is denoted by extracellular matrix accumulation, renal fibrosis, and cellular mitogenesis (41). In creases in protein levels of TGF, fibronectin, and collagen IV are important ear ly markers of these processes (42,43). The evidence for the involvement of a Jak2/ROS inter action in diabetic nephropathy is compelling. Jak2 has been shown to be responsible for increa ses in these proteins in glomerular mesangial

PAGE 24

24 cells in response to hyperglycemia (42,43). Furthermore, hyperglycemia-enhanced Jak2 activation has been shown to increase cell proliferat ion in glomerular mesangial cells (43). Thus, Jak2 has been implicated in the key transcrip tional and proliferative events of diabetic nephropathy. The notion that these events are ROS-dependent is strongly suggested in the literature. The hyperglycemia-induced ROS mechanisms that play prominent roles in the development of diabetic nephropathy are the polyol pathway a nd Advanced Glycation Endproducts (AGEs). AGEs are important indicators of cellular oxidative st ress levels. Not only is AGE production facilitated by ROS, but AGEs can further incr ease ROS levels thr ough interactions with scavenger receptors. AGEs were found to increase TGF, fibronectin, and collagen IV synthesis in glomerular mesangial cells (44). Furthermore, these increases were found to be ROS-dependent as they were inhi bited by the antioxidant, vitamin E succinate (44). Treatment of mesangial cells with an aldose reductase inhibitor effectively blocked AGE-mediated increases in TGFand collagen IV levels, strongly suggesti ng an interplay between AGEs and the polyol pathway (45). It was also shown that the AGE-d ependent increases in TGFand fibronectin are mediated by the AT1 receptor, a known Jak2 activator (4 6). Finally, it was shown that AGEinduced Jak2 activation led to increased cellular mitogenesis in renal fibr oblasts, an aggravating factor for diabetic nephropathy (47). In summary, Jak2 has been implicated in the major gene expression events associated with diabetic nephropathy. Hyperglycemia provides th e link between Jak2 and ROS in this disorder. The hyperglycemic condition results in ROS pr oduction through increased polyol pathway flux and production of AGEs. These two ROS sources ha ve been linked to the same gene expression events that Jak2 has been implicated in. AGE effects have been shown to be mediated by the

PAGE 25

25 AT1 receptor, a well-known upstream element of Jak-STAT signaling. Finally, AGEs have been shown to mediate Jak2 activation in renal fibroblasts and further aggravate diabetic nephropathy through increased cellular proliferation. Thus, the evidence linking a Jak2/ROS interaction to the progression of diabetic nephropathy is very su bstantial. While the current literature links Jak-STAT signaling to ROS sources like AGEs and the polyol pathway, there have been no direct links between ROS levels and Jak2 activation in diabetic nephropathy. Studies utilizing antioxidant therapy in attempts to block Ja k2-dependent increases in TGF-, collagen IV, fibronectin, and cell proliferati on associated with diabetic ne phropathy would be excellent additions to this field. Diabetic cardiomyopathy Diabetic complications are quite pervasiv e and can develop in many different organ systems. While diabetic nephropathy has se vere health conseque nces, cardiovascular complications produce the bulk of the mortality associated with diabetes mellitus. Diabetic cardiomyopathy is a one such complication. Th is disorder is characterized by a loss of ventricular function, cardiac hypertrophy, and cardi ac fibrosis (48). Ul timately, the progression of this disease results in heart failure. Hyperglycemia-induced oxidative stress play s a central role in the development and progression diabetic cardiomyopathy. Hyperg lycemia-induced ROS produc tion has been linked to cell death in both cardiomyocytes and cardiac stem cell populations, re ducing the contractile function of the heart and its regenerative cap acity (48,49). ROS-dependent signal transduction events that contribute to di abetic cardiomyopathy include the upregulation of scavenger receptors for AGEs, reduced levels of the cardioprotective transc ription factors HAND and MEF-2, and a shift in myosin heavy chain isoforms from alpha myosin heavy chain to the beta

PAGE 26

26 myosin heavy chain (50). The endogenous anti oxidant dehydroepiandros terone effectively suppressed these changes (50). In contrast to diabetic nephr opathy, diabetic cardiomyopathy is a disease in which the role of the Jak2/ROS relationship is much less clear. It has been shown that hyperglycemia-induced ROS production activates Jak2 in failing cardiomyocytes in an AT1 receptor-dependent manner (51). Furthermore, high glucose was determined to be a necessary prerequisite for angiotensin II-dependent Jak2 activation in non-failing myocyt es (51). However, when investigating the functional alterations of cardiomyoc ytes in diabetic cardiomyopathy, such as increases in time to peak shortening (TPS) and time to 90% relengt hening (TR90), it was found that Jak2 inhibition via AG490 did not affect these increases (52). Co llectively, these lines of evidence suggest that ROS-mediated Jak2 activation may not play a ro le in the early deve lopment of diabetic cardiomyopathy. Rather, this mechanism may only become a significant factor in advanced cardiomyopathy, when cardiomyocytes have begun to fail. Atherosclerosis Atherosclerosis has become a major health concern in developed countries. This cardiovascular disorder is characterized by the pr esence of arterial lesions (53). These lesions often develop into fibrous plaque s that protrude into the arteri al lumen (53). The plaques are composed of a number of different cell types, including vascular smooth muscle cells and lipidpacked macrophages known as foam cells (53). Atherosclerotic plaques that become unstable may rupture and release their contents into the vessel lumen (53). This event can lead to the formation of blood clots. In th e coronary circulati on, these clots can impede blood flow, causing an acute myocardial ischemic event (53). There is an abundance of evidence that suggests that Jak2 plays a ROS-dependent, facilitative role in the progression of atherosclero sis. The complexity of this role is unique

PAGE 27

27 because it relies on both cellula r proliferation and apoptosis to achieve the progression of atherosclerosis. Vascular smoot h muscle cell proliferation is a necessary component of early atherosclerotic development (53). ROS have been shown to signifi cantly induce vascular smooth muscle cell proliferation (54). In a st udy utilizing hydrogen pero xide as a source of oxidative stress, it was shown that vascular sm ooth muscle cells responded to ROS with an increase in activity of the proliferative ERK2 signaling pathway (55). Furthermore, ROS also induced an increase in the expression of Heat Shock Protein 70 (HSP70), a protein that possibly contributes to the proliferativ e phenotype by countering the proapoptotic process of protein aggregation (56). The Jak2 inhi bitor, AG490, achieved partial blockade of ERK2 activation and complete inhibition of HSP70 induction, suggesting that Jak2 plays a signifi cant role in this proliferative mechanism (56). Thrombin, a known mitogen for vascular sm ooth muscle cells, elic its its proliferative effect through the same mechanism descri bed for that of hydrogen peroxide (56). Pharmacological Jak2 inhibition reinforced the role of Jak2 in ERK2 activation and HSP70 expression (56). Furthermore, Ja k2 inhibition was directly linked to a complete eradication of a thrombin-mediated increase in vascular smooth muscle cell proliferation (56). Thrombininduced Jak2 activation was substantially inhi bited by a number of different antioxidants, strongly suggesting that ROS mediate thrombin-indu ced Jak2 activation (56) Angiotensin II is another potent mitogen for vasc ular smooth muscle cells. It was shown that the antioxidant caffeic acid reduced angiotensi n II dependent increases in superoxide production and proliferation in vascular smooth muscle cells (57). This compound also blocked angiotensin IIdependent Jak2 activation, suggesting that angiot ensin II affects vascular smooth muscle cell proliferation through ROS-dependent Jak2 activati on (57). Collectively, these lines of evidence

PAGE 28

28 strongly suggest that Jak2 plays a critical ro le in mediating ROS-de pendent vascular smooth muscle cell proliferation. Although important, increased vascular smooth mu scle cell proliferation does not tell the entire story of atherosclerosis. Apoptosis, or programmed cell death, is al so an important factor in the progression of this diseas e. It is thought that vascular smooth muscle cell apoptosis in mature atherosclerotic plaques contributes to plaq ue instability (58). Unstable plaques are more likely to rupture, spilling their thrombotic contents into the arterial lumen (58). In an interesting contrast to the evidence above, oxi dative stress has also been shown to be an apoptotic factor for vascular smooth muscle cells (58). Specifica lly, it was shown that oxidized low-density lipoprotein induced ROS production and apoptosis in vascular smooth muscle cells (58). Antioxidant treatment ameliorated the apoptotic e ffect (58). A separate study demonstrated that oxidized low-density lipoprotein ca n activate Jak2, suggesting that th e effect seen in vascular smooth muscle cells may be Jak2-dependent (59). Our lab elucidated a direct link between Jak2 activation and oxidative stress-d ependent apoptosis in vascular smooth muscle cells (60). Specifically, it was shown that Jak2 inhibition ab olished hydrogen peroxide-induced apoptosis in vascular smooth muscle cells (60). Finally, Jak2 was also linked to ROS-dependent apoptosis in aortic endothelial cells, suggesti ng that the Jak2/ROS interacti on may also play a role in endothelial dysfunction, an important in itiating event of at herosclerosis (61). In summary, the current literature attributes a pathogenic role to th e Jak2/ROS interaction in atherosclerosis. Jak2 appears to be a critical factor in both th e pro-apoptotic and proliferative effects of ROS. Interestingly, Jak2 mediated the detrimental effects of RO S at both the early and late stages of atherosclerosis, as well as in different cell types. These findings suggest that Jak2 may be an important therapeutic target in the treatment of atherosclerosis.

PAGE 29

29 Cardiac Ischemia-Reperfusion Injury Cardiovascular disease is a serious health co ncern in the developed world. Within this disease category is myocardial Is chemia/Reperfusion (I/R) injury. This insult involves the loss and subsequent restoration of coronary blood fl ow to the myocardium. The loss of coronary blood flow may result from atherosclerotic lesions, blood clots, or surgical procedures like heart transplants. While restoring bl ood flow to the myocardium is clearly necessary, reperfusion is also thought to be the trigger for much of the apoptosis and necrosis that is characteristic of I/R injury. The prevailing view in the I/R field is that reperfusion causes significant ROS production and that this oxidative stress is a si gnificant mediator of I/R injury. The importance of ROS in I/R in jury is undeniable and well-doc umented in the literature. In a rat model of I/R injury, administration of th e antioxidant dioclein prior to the induction of I/R injury significantly reduced ROS levels afte r I/R injury and preven ted the development of post I/R arrythmias (62). In a nother study, the antioxidant lycopene was shown to be effective at restoring Mean Arterial Pressure and Heart Ra te after I/R injury (63). Resveratrol, an antioxidant found in red wine, was found to reduce infarct size after I/R injury (64). Finally, Vitamin C was shown to inhibit ROS productio n and apoptosis in a hypoxia-reoxygenation model of I/R injury (65). The relationship between ROS and Jak2 in the c ontext of I/R injury is very complex. The I/R literature contains several examples in which Jak2 serves a cardi oprotective function. The phenomenon of Ischemic Preconditioning (IPC), in which the myocardium is protected by exposure to several short ischemic episodes prior to a larger ischemic event, is a prominent example of a Jak2-dependent, ca rdioprotective mechanism. In a mouse model of I/R injury, ischemic preconditioning was found to incr ease Jak2, STAT1, and STAT3 phosphorylation levels (66). Furthermore, pharmacological blockade of Jak-STAT signaling via AG490 was

PAGE 30

30 shown to eliminate the IPC-dependent reduction in infarct size (66). In addition to being Jak2dependent, IPC is also ROS-depe ndent as its benefi cial effects on the myocardium can be inhibited by antioxidant pretreat ment (67). With the knowledge that Jak2 can be activated by ROS (11), these findings suggest that the fr ee radicals produced by IPC activate Jak-STAT signaling to produce a cardi oprotective phenotype. In addition to IPC, there are also exampl es of pharmacological preconditioning in which Jak2 may interact with ROS. In an isolated ra t heart model of I/R, it was found that pre-treating the hearts with TNFresulted in a reduction in infarct si ze (68). Furthermore, proapoptotic proteins, such as the Bcl-2 agonist of cell death (BAD) we re inhibited by TNFtreatment (68). This protective mechanism was shown to be STAT3-dependent and the effects of TNFpreconditioning were blocked by the Jak-STAT inhi bitor, AG490 (68). In a separate study, free radicals were shown to play a critical role in TNFpreconditioning (69). Thus, these studies bolster the link between ROS and Jak2 in the development of a cardioprotective phenotype. Angiotensin II preconditioning is characterized by effects sim ilar to those seen in TNFpreconditioning. Reductions in infarct size and ca rdiomyocyte apoptosis were both seen as a result of angiotensin II preconditioning in an isol ated rat heart model (70). Furthermore, these effects were shown to be dependent on RO S production from the membrane-bound enzyme, NADPH oxidase (70). In a sepa rate study, it was shown that th e angiotensin II-NADPH oxidase pathway precedes Jak2-dependent activation of STAT1 and STAT3 in rat aortic smooth muscle cells (71). Therefore, it is lik ely that Jak2, through its interact ion with ROS, plays a role in angiotensin II preconditioning. Jak2 has been linked to several other car dioprotective pathwa ys, including opioiddependent cardioprotection, IL-6 preconditioning, and granulocyt e colony stimulating factor-

PAGE 31

31 dependent cardioprotection (7274). A common thread among all of the pathways mentioned above is the Jak2-dependent activation of STAT3 (66,68,70,71-74). This commonality adds another dimension to the Jak2/RO S relationship in I/R injury. While ROS may activate Jak2 in several cardioprotective pathways Jak2 also appears to play a role in reducing myocardial oxidative stress through STAT3 activation. STAT3 activation has been linked to increases in antioxidant proteins like metallo thionein and manganese superoxide dismutase in response to I/R injury and hypoxia/reoxygenation, respectively (75,76). In both cases, these increases led to significant reductions in intracellular ROS leve ls (75,76). In the case of metallothionein upregulation, the STAT3 dependent increase in th is protein led to a si gnificant reduction in infarct size (75). Therefore, Jak2 may offer a significant level of protection through STAT3 activation. Furthermore, with respect to the ROS-generating pathways (IPC, TNF, and angiotensin II), Jak2 could potentially mediate a negative feedback loop in which increased oxidative stress leads to enha nced antioxidant activity. The final twist in this Jak2/ROS interacti on results from the finding that Jak2 does not always function in a cardioprotective manner. It was recently shown that, in response to hypoxia-reoxygenation, Jak2 activation resulted in a significant increase in cardiomyocyte apoptosis (77). Furthermore, Jak2 activation in response to h ypoxia-reoxygenation was inhibited by the antioxidant dehydroascorbic acid, sugges ting that oxidative stre ss is the activating stimulus (65). This inhibition coincided with a decrease in cardiomyocyte apoptosis, bolstering the assertion that cardiom yocyte apoptosis in I/R inju ry is Jak2-dependent (65). In summary, the relationship between Jak2 and oxidative stress in the context of I/R injury is anything but straightforward. Jak2 is an integral compone nt of several cardioprotective pathways. There is evidence to suggest that Ja k2 activation mediates increased ROS scavenging

PAGE 32

32 and is a necessary component in a number of ROS-dependent negative feedback loops. However, the evidence linking Jak2 activation to ROS-dependent card iomyocyte apoptosis stands in sharp contrast to the cardioprotective hypothesi s. The reconciliation of these two roles may lie in the relationship between ROS and I/R in jury. While it is true that oxidative stress appears to be the main mediator of I/R injury, th e studies discussed above also assert that free radicals can be beneficial in smaller doses. In a study using a rabbit heart model of I/R injury, low doses of ROS given prior to the onset of I/R were determin ed to reproduce the cardioprotection afforded by IPC, suggesting th at lower amounts of ROS are released by IPC than longer ischemic bouts (78). Furthermore, ceramide, a molecule known to increase oxidative stress, has been found to exacerbate I/R injury in large doses and precondition the myocardium in low doses (79). This example bolsters the assertion that ROS pl ay a dual role in I/R injury. Interestingly, Jak2 activation appear s to parallel the role of ROS in I/R injury in that mechanisms that generate smaller doses of ROS, such as IPC, elicit the ca rdioprotective effects of Jak2 activation. On the other hand, longer ischemic bout s, which presumably generate higher levels of ROS, evoke the detrimental effects of Jak2 ac tivation. Thus, the Jak2/R OS interaction in I/R injury may be one in which Jak2 responds to the dose of ROS and not just the mere presence of oxidative stress. This differentia l activation would be an interesti ng area for future studies in the I/R field.

PAGE 33

33 Figure 1-1. Jak Homology (JH) dom ains. All four members of the Janus family share seven structural features in common. They ar e termed Jak Homology domains. The cterminal JH1 domain is the site of kinase activity. The JH2 domain is known as the pseudokinase domain and has been shown to have an inhibitory effect on the JH1 domain. The JH3 and half of the JH4 do main encode a putative SH2 domain. The FERM domain extends from the second half of the JH4 domain to the JH7 domain and facilitates Jak/ receptor interaction.

PAGE 34

34 CHAPTER 2 METHODS In Silico Molecular Modeling of Jak2 The Swiss Model program was used to genera te a structure homology model of murine Jak2 based on the human Jak2 crystal structure (P DB code: 2B7A). The Definition of Secondary Structure of Proteins (DSSP) pr ogram was used to calculate th e solvent accessible surface areas of tyrosine 972 and serine 1120 (80). Mass Spectrometry Using a vaccinia virus overexpression syst em, wild-type Jak2 protein was overexpressed in BSC-40 cells and purified as previously de scribed (81). The purified protein was then separated by SDS-PAGE, coomma ssie stained, excised from the gel and subjected to ms/ms mass spectrometry as prev iously described (81). Cell Lines Three cell lines were used to carry out thes e investigations; COS-7 cells, BSC-40 cells, and 2A/GHR cells. The COS-7 and BSC-40 cell li nes both originate from the monkey kidney. Both of these cell lines exhibit very low endoge nous Jak2 expression. Thus, they both provide suitable backgrounds on which to carry out transien t transfection studies. The BSC-40 cells are more permissive to vaccinia virus infection than COS-7 cells. Thus, they were used in favor of COS-7 cells for all overe xpression assays. The 2A/GHR cells are mouse fibroblast cell that have been subjected to gamma irradiation. Thes e cells contain non-functional Jak2 alleles. Our laboratory has stably transfected these cells w ith the growth hormone receptor. Thus, they provide an excellent platform on which to carry out growth hormone signa ling studies since they stably express the growth hormone receptor.

PAGE 35

35 Cell Culture BSC-40 cells were grown in high glucose (4.5 g/L) DMEM supplemented with 10% newborn calf serum. COS-7 cells were grown in high glucose DMEM supplemented with 10% fetal bovine serum. 2A cells that stably express the growth hormone receptor were cultured in low glucose (1g/L) DMEM supplemented with 10 % fetal bovine serum, neomycin (0.2 mg/ml), and zeocin (0.1 mg/ml). All cells were cultured at 37C and 5% CO2. Cells treated with angiotensin II or growth horm one were growth-arrested with serum-free DMEM for 18 hours prior to treatment. Site-Directed Mutagenesis pRC-CMV Jak2 Y972F, pBOS Jak2 Y972F, pRC-CMV Jak2 S1120A, and pBOS Jak2 S1120A were all made using the Stratagene Qu ikChange Mutagenesis protocol. The sense primer sequence for the Jak2 Y972F plasmids is 5'CTTGGTACAAAAAGGTTTATCCAC AGGGACCTG. The antisense primer sequence for these plasmids is 5'-CAGGTCCCTGTGG ATAAACCTTTTTGTACCAAG. For the Jak2 S1120A plasmids, the sequence for the sense primer is 5CTTCAGGGACCTTGCGTTCGGGTGGATC A. The sequence for the antisense primer is 5TGATCCACCCGAACGCAAGGTCCCTG AAC. DNA sequencing was used to verify all mutations. Transient Cell Transfections For each transfection, plasmid DNA and lipofec tin were incubated in separate 0.5 ml aliquots of serum-free DMEM at room temper ature for 0.5 hours. Plasmid DNA and lipofectin were then combined and incubated at room temp erature for 10 minutes. During this incubation, the cells to be transfected were washed twice with PBS. An additional 2 ml of serum-free DMEM was added to each DNA/Lipofectin solution and the 3 ml transfection mixture was then

PAGE 36

36 pipetted onto a plate of cells. A ll cells were returned to 37C incubation for five hours. After five hours, transfection mixtures we re then aspirated off of the cel ls and replaced with 5 ml of serum-containing DMEM and the cells we re allowed to recover overnight. Immunoprecipitation Cells were washed twice with ice-cold PB S containing 1mM sodium orthovanadate. The cells were then lysed with 900 ul of ice-cold RIPA buffer containing pr otease inhibitors and placed on ice. Cellular lysates were harvested and briefly sonicated at 3.3 Hz. The sonicated lysates were put on ice for 1 hour. Lysates were then centrifuged at 16000 g for 5 min. The supernatants were transferred to new tubes and the pellet wa s discarded. Whole cell lysate samples were prepared by adding a 50 ul sample of each lysate to 15 ul of 4X SDS sample buffer. The remainder of each lysate was used for immunoprecipitation. Twenty microliters of Protein A/G beads (Santa Cruz Biotechnology) and 2 ug of the appropriate antibody were added to each lysate. All immunoprecip itations were incubated at 4C with shaking for 4-18 hours. After incubation, immunoprecip itations were centrifuged at 7000 rpm for 2 min. The supernatants were discarded and the pellets were washed with 1 ml of IP wash buffer, three times. The protein A/G beads were then resusp ended in 65 ul of 1X SDS sample buffer. Immunoprecipitated proteins and whole cell ly sate samples were separated on an 8% polyacrylamide gel for 1200 volt-hours. The se parated proteins were then transferred to nitrocellulose membranes for 300 volt-hours. Western Blotting All western blots were executed at room temp erature. Nitrocellulose membranes were blocked in 30 ml of either 5% BSA/TBST or 5% milk/TBST for 1 hour. The membranes were then incubated in 25 ml of primary antibody so lution for 1-2 hours and washed in TBST for 1 hour. Membranes were incubated in secondary antibody solution (1:4000, GE Healthcare) for 30

PAGE 37

37 min and then washed in TBST for 30 min. Proteins were visualized via enhanced chemilluminescence reagents. Luciferase Assay COS-7 cells were transiently transfected in the manner described above with the appropriate plasmid DNA and 2 ug of a luciferase reporter constr uct downstream of four tandem copies of the interferonactivating sequence (pLuc-GAS). Af ter 5 hours in transfection media, all cells were trypsinized and seeded into six well culture plates. For ligand-independent gene expression experiments, 7 x 105 cells were seeded into each well. For growth hormone dependent gene expression experiments, 2 x 105 cells were seeded into each well. The cells were allowed to recover in DMEM serum media for 24 hours. For growth hormone dependent gene expression experiments, all cells were serum star ved for 16 hours. All cells were then lysed in 1X Reporter Lysis buffer (Promega) for a minimum of 5 hours. During this lysis period, the lysates were subjected to one freeze-thaw cycle between room temperature and -80C. A 20 ul sample of each lysate was mixed with 100 ul of lu ciferase substrate and re lative light units were read by a Monolight 3010 luminometer. Fluorescence Microscopy 2A cells stably expressing the growth hormone receptor were cultured on microscope slides. These cells were transiently transfect ed with 10 g of a plasmid encoding a Green Fluorescent Protein-tagged STAT1 protein. Additionally, these cells were co-transfected with 10 g of either Jak2 wild-type plasmid or Jak2 Y972F plasmid. The cells were then treated with 250 ng/ml growth hormone. The cells were washed once with 1X PBS and then they were fixed with 200 l of 4% paraformaldehyde for 10 minutes at room temperature. The cells were then washed three times in 1X PBS and allowed to dry. The slides were then mounted with

PAGE 38

38 VectaShield and DAPI mounting solution. Coverslips were then placed ove r the slides and the slides were viewed under a conf ocal microscope. The fluorescen ce intensity of GFP-STAT1 was quantified using ImageJ software.

PAGE 39

39 CHAPTER 3 Y972 AND JAK2 ACTIVATION1 Summary Janus Kinase 2 is a 130 kDa tyrosine kinase that transduces gene transcription signals from the cell surface to the nucleus. Its func tion can be intrinsically regulated by the autophosphorylation of a handful of its 49 tyrosines. Here, we describe the characterization of tyrosine 972 (Y972), a novel Jak2 regulatory site. Through ms/ms mass spectrometry, we found that Y972 is a site of Jak2 autophosphorylation. We used site-directed mutagenesis to introduce a Y-F point mutation at position 972 in plasmids encoding wild -type Jak2 protein. Using these plasmids, we investigated the consequences of losing phosphorylation at Y972 on Jak2 function. We determined that the loss of Y972 phosphoryl ation significantly reduc ed both Jak2 total tyrosine phosphorylation and Y1007/ Y1008 phosphorylation. Additionally, Y972 phosphorylation was shown to be important for maximal Jak2 kinase function. Surprisingly, Y972 phosphorylation did not affect ligand-independent gene expr ession. Finally, the loss of Y972 phosphorylation impaired severa l aspects of Jak2-dependent si gnal transduction, including angiotensin II-dependent Jak2 phosphoryla tion, growth hormone-dependent Y1007/Y1008 phosphorylation, and growth hormone-dependent ST AT1 nuclear transloca tion. Collectively, the data suggest that we have identified a novel site of Jak2 f unctional regulation. Introduction Janus Kinase 2 (Jak2) is one of four family members of the Janus family of tyrosine kinases. These proteins mediate signals from the cell surface to the nucleus through tyrosine phosphorylation signaling cascades The primary cellular role of Jak2 is to phosphorylate members of the Signal Transducers and Activators of Transcription (STAT) family of latent 1 Reprinted with permission from the American Chemical Society and Drs. Peter P. Sayeski and David Ostrov..

PAGE 40

40 cytoplasmic transcription factors. Once phosphoryl ated, STAT proteins can then dimerize and translocate to the nucleus. In the nucleus, STAT complexes bind DNA promoter elements and alter cellular gene transcription patterns. The rele vance of Jak2 to a wide array of disease states highlights the fact that Jak2 f unction must be exquisitely re gulated (82). Jak2 functional regulation is achieved through th e cooperation of several differe nt extrinsic and intrinsic elements. Extrinsically, Jak2 function is regulated by a multitude of proteins that combine to produce a specific Jak/STAT signaling pathway. These proteins include cell surface receptors, adaptors/activators, and negative regulators. On an intrinsic level, Jak2 is regulated by the phosphorylation of several of its 49 tyrosine residues. Currently, a handful of studies have defined regulatory mechanisms that are associated with Jak2 tyrosine phosphoryla tion (19-24). Tyrosine 1007 resi des in the Jak2 activation loop and its phosphorylation is required for maximal ac tivation of Jak2 (19). Our lab has shown that phosphorylation at Y201 is necessary for the inte raction of Jak2 with SHP-2, an important adaptor protein that plays a role in angiotensi n II-dependent Jak2 activation (23). Tyrosines 221 and 570 have been shown to modulate Jak2 kinase function (22). Given the importance of these phosphotyrosines to Jak2 functi on, the study of new tyrosine phos phorylation sites may progress our understanding of Jak2 function. The extrinsic components of Jak2 functional regulation include the cell surface receptors and adaptor proteins that help propagate a Jak/ STAT signal to the nucleus. Since Jak2 can be activated by both G-protein coupled receptors (GP CRs) and cytokine receptors, Jak2-dependent signaling can be broken up into two main paradigms. Of the two paradigms, the cytokine model is more fully understood. In the cytokine model, Jak2 is constitutively bound to the cytoplasmic portions of the receptor subunits (25,26). While it was originally thought that ligand binding at

PAGE 41

41 the cell surface led to re ceptor subunit dimerization, recent evidence suggests that ligands bind pre-dimerized receptor subunits (27). Ligand/re ceptor association is thought to initiate conformational changes in the receptor subunits that ultimately place the constitutively bound Jak2 molecules in close enough proximity to one another to achieve tran sphosphorylation (27). Once activated, Jak2 phosphorylates the cytoplasmic tail of the re ceptor to produce recruitment sites for STAT proteins (27). The recruited ST AT proteins are then phosphorylated by Jak2 and the subsequent events occu r as described above. The GPCR model differs from the cytokine mode l at a few notable points. First, it is thought that GPCRs activate a cytoplasmic pool of Jak2 proteins. Additionally, there is currently no evidence to suggest that GPCRs facilitate the same level of physical contact between Jak2 molecules that cytokine receptors are thought to en able. Rather, it is thought that, upon receptor activation, the G q receptor subunit activates Jak2 (29). On ce activated, Jak2 then translocates to the cytoplasmic tail of the GPCR (83). Phosphor ylated tyrosines within Jak2 are thought to serve as STAT recruitment sites (84). Once recruited, STAT proteins are phosphorylated by Jak2 and the subsequent events are th e same as those described above. The Src Homology 2 B (SH2B) protein family makes up another important component of Jak2 extrinsic regulation. This protein family consists of three member s; SH2B, APS, and Lnk (34). All members have both an SH2 domain a nd a pleckstrin homology domain (34). There are multiple isoforms of the SH2B protein (34). The SH2Bisoform is known to bind Jak2 at phosphorylated tyrosine 813 ( 21). Upon binding, SH2Bsignificantly enhances Jak2 activation, particularly in response to growth hormone (21). It is thought that SH2Bcan facilitate Jak2 dimerization and stabilize the active Jak2 conforma tion (35). These two molecular processes are thought to assist Ja k2 autophosphorylation (35). In this study, we

PAGE 42

42 characterized Y972, a novel site of Jak2 autopho sphorylation. Phosphorylation at Y972 was determined to be critical for the maintenance of both total tyrosine phosphorylation levels and Y1007 phosphorylation. Furthermore, Y972 phosphoryl ation differentially affects several Jak2dependent signal transduction mechanisms, su ch as growth hormone and angiotensin IIdependent Jak2 phosphorylation. As such, this wo rk has identified a novel mechanism of Jak2 tyrosine kinase regulation. Results Tyrosine 972 is a Site of Jak2 Autophosphorylation BSC-40 cells were transiently transfected with 5 g of Jak2 wild-type plasmid. Using a vaccinia virus-mediated overexpression system, Ja k2 wild-type protein was expressed at a high level and purified in a manner previously described (81). This purified Jak2 protein was subjected to ms/ms mass spectrometry. The re sults suggest that Y972 is a site of Jak2 autophosphorylation (Fig. 1). Th e phosphorylation statuses of Y221 and Y1007, two previously characterized phosphotyrosines (19,22) were verified as controls. Tyrosine 972 is Solvent Accessible We performed in silico modeling of the Jak2 kinase domain to determine the solventaccessible surface area of Y972. Using the Sw iss model program, a structure homology model of the murine Jak2 kinase domain was generated based on the crystal structure of the human Jak2 protein (protein database code: 2B7A). Using th e Definition of Secondary Structure of Proteins (DSSP) program (80), we determined that the solvent accessible surface area of Y972 is 26 2. As 15 2 is generally considered the minimum su rface area required for a phosphorylation site, the solvent accessible surface ar ea of Y972 is conducive to phosphorylation. We have shown a representation of the Jak2 ki nase domain with Y972 highlighted in red (Fig. 2).

PAGE 43

43 Y972 Affects Jak2 Total Tyrosine Phos phorylation and Y1007 Phosphorylation Previous studies have shown that phosphotyros ines can play important roles in regulating Jak2 autophosphorylation levels (19-24). To de termine the effect of Y972 on total Jak2 phosphorylation, empty vector, Jak2 wild-type, a nd Jak2 Y972F plasmids were overexpressed in BSC-40 cells. Jak2 protein was isolated via immunoprecipitation and total tyrosine phosphorylation was measured via anti-phospho tyrosine western blot. The loss of phosphorylation at Y972 dramatically reduced Jak2 total tyrosine phosphoryl ation (Fig. 3A). A similarly severe reduction in Y1007 phosphoryl ation was seen in response to losing phosphorylation as compared to Jak2 wild-type protein (Fig. 3C). In both cases, a basal level of phosphorylation still remained in both cases even though significant reductions were seen. Thus, the results suggest that the loss of Y972 phosphorylation aff ects both Jak2 total and Y1007 phosphorylation. The membranes were stripped a nd equal protein loading was determined via western blot (Figs. 3B and 3D). The Loss of Y972 Phosphorylation Affects Jak2 Kinase Activity, But Does Not Confer a Dominant Negative Phenotype. One possible explanation for the dramatic lo sses in phosphorylation seen above is that the loss of Y972 phosphorylation confers an inhi bitory phenotype to Jak2. Structural and conformational changes w ithin the Jak2 kinase domain can generate such a phenotype (14,15). To evaluate the inhibitory potential of Ja k2 in the absence of Y972 phosphorylation, BSC-40 cells were transfected with 5 g of a plasmi d encoding an HA-tagged Jak2 wild-type protein. Additionally, these cells were co -transfected with increasing am ounts (5 and 15 g) of plasmids encoding either Jak2 wild type protein, a Jak2 dominant negative prot ein (W1020G/E1024A), or Jak2 Y972F protein. These plasmids were overexpressed using the vaccinia virus overexpression system (81). The cells were lysed and HA-tagged Jak2 was immunoprecipitated

PAGE 44

44 from the lysates. The levels of Y1007 phosphoryl ation of the HA-tagged Jak2 wild type protein were determined by western blot. The Jak2 Y 972F mutant did not inhi bit HA-tagged Jak2 wildtype phosphorylation, but it also did not increase wild-type ph osphorylation (Fig. 4A). In contrast, Jak2 wild-type pr otein increased Y1007/Y1008 phosphor ylation levels in HA-tagged Jak2 wild-type protein and the Ja k2 dominant negative protein reduc ed these levels (Fig. 4A). The membrane was stripped and re-probed with an anti-HA polyclonal antibody to verify equal protein loading (Fig. 4B). In a similar experiment, we evaluated the abilit y of Jak2 to act as a s ubstrate in the absence of Y972 phosphorylation. BSC-40 cells were tr ansfected with 10 g of either an HA-tagged Jak2 wild type plasmid or an HA-tagged Jak2 Y972F plasmid. Additionally, these cells were cotransfected with increasing amount s (5 and 15 g) of Jak2 wild type plasmid. These plasmids were overexpressed using the vaccinia virus ove rexpression system (81). The levels of Y1007 phosphorylation in the HA-tagged Jak2 isoforms we re measured as described above. It was determined that the loss of Y972 phosphor ylation did not prevent Jak2 from being phosphorylated at Y1007 by a wild-type Jak2 mo lecule (Fig. 4C). In summary, these experiments demonstrate that the loss of Y972 phosphorylation reduces Jak2 kinase function but does not prevent Jak2 from being phosphorylated at Y1007. For both experiments, equal protein loading was determined via wester n blot (Figs. 4B and 4D). Finally, we looked at the e ffect of Y972 phosphorylation on Jak2 dimerization. COS-7 cells were transfected with 10 g of either an empty vector plasmid, a FLAG-tagged Jak2 wildtype plasmid, or a FLAG-tagged Jak2 Y972F pl asmid. Additionally, these cells were cotransfected with 10 g of an HA-tagged Ja k2 wild-type plasmid. Forty-eight hours after transfection, all cells were lysed and HA-tagge d Jak2 protein was immun oprecipitated from the

PAGE 45

45 lysates with an anti-HA monoclonal antibody. Th e levels of co-precipitated FLAG Jak2 protein were determined via western blot. The Y972F mutation did not affect the ability of FLAG-Jak2 to co-precipitate with HA-Jak2 wild-type pr otein, suggesting that Jak2 dimerization is independent of Y972 phosphorylation (Fig. 4E). The membrane was then stripped and the levels of HA-tagged Jak2 protein were determ ined via western blot (Fig. 4F). Tyrosine 972 Does Not Affect Liga nd-Independent STAT1 Activation. Having examined how Y972 affect s the intrinsic biochemistry of Jak2, we then focused our attention on how Y972 phosphorylation affects extrinsic aspects of the Jak/STAT pathway, such as STAT1 activation. COS-7 cells were tr ansiently transfected w ith increasing amounts of either Jak2 wild-type or Jak2 Y972F plasmi d. The cells were lysed and STAT1 was immunoprecipitated from the lysates. STAT1 phosphorylation at Y701 was determined via western blot. It was determined that the loss of phosphorylation at Y972 did not hinder the ability of Jak2 to phosphorylate Y701 in STAT1 (F ig. 5A). The membrane was stripped and reprobed to verify equal prot ein loading (Fig. 5B). Tyrosine 972 Phosphorylation Does Not Aff ect Ligand-Independent Gene Expression Even in the absence of ligand treatment, Jak2 is capable of driving a basal level of gene expression (85). This ligand-independent gene expression represents the enzymes intrinsic functional capacity. A luciferase gene repor ter assay was performed to test whether the phosphorylation status of Y972 affects this in trinsic functional capacity. COS-7 cells were transiently transfected with a plasmid encoding four tandem repeats of the -activating sequence upstream of the firefly luciferase protein. Addi tionally, these cells were co-transfected with increasing amounts of plasmid encoding either Ja k2 wild-type protein or Jak2 Y972F protein. These cells were lysed and lysates were treated with an excess of luciferase substrate. Relative light units were taken as a measure of Jak2-depe ndent gene expression. There was no significant

PAGE 46

46 difference between the abilities of Jak2 wild type protein a nd Jak2 Y972F protein to drive luciferase gene expression (Fig. 6A). A portion of the transfected ce lls were used to verify equal expression levels of Jak2 wild-type and Jak2 Y972F via western blot (Fig. 6B). Tyrosine 972 is Important for the Jak2 Response to Angiotensin II Having investigated the impact of Y972 phosphorylation on ligand-independent Jak2 function, we then decided to evaluate the rele vance of Y972 phosphorylat ion for Jak2 function in a signaling context. One of the Jak2-dependent signal transduction pathways is the angiotensin II pathway. Given the physiological importance of this pathway, we sought to determine the effect of Y972 phosphorylation on th e ability of Jak2 to respond to angiotensin II. COS-7 cells were transfected with plasmids expressing e ither the angiotensin type 1 receptor (AT1R) and Jak2 wild-type protein, or the AT1R and Jak2 Y972F protein. Thes e cells were treated with angiotensin II and then lysed. Jak2 protein was immunoprecipita ted from the lysate and Jak2 phosphorylation in response to treatment was measured via western blot. The loss of phosphorylation at Y972 abolished the angi otensin II-dependent increase in Jak2 phosphorylation (Fig. 7A). The membrane was stri pped and re-probed to verify equal protein loading (Fig. 7B). Tyrosine 972 has Differential Effects on Growth Hormone-Dependent Jak2 Total and Y1007 Phosphorylation. The growth hormone signaling pathway is an excellent example of the cytokine model of Jak2-dependent signaling. Thus, we sought to determine the impact of Y972 phosphorylation on the growth hormone signaling pathway. COS-7 cells were transiently transfected with plasmids encoding either the growth hormone receptor and Jak2 wild type protein, or the growth hormone receptor and Jak2 Y972F protein. These cells were treated with growth hormone and then lysed. Jak2 protein was immunoprecipita ted from the lysate and the Jak2 phosphorylation response to

PAGE 47

47 growth hormone was determined via western blot The loss of Y972 ph osphorylation severely hindered growth hormone dependent increases in Y1007/Y1008 phosphorylation (Fig. 8A). The membrane was stripped and reprobed with an anti-Jak2 polyclonal antibody to verify equal protein loading (Fig. 8B). Growth hormone de pendent total tyrosine phosphorylation was also investigated via western blot. It was determined that the loss of phosphorylation at Y972 did not significantly reduce the growth hor mone dependent increase in to tal phosphorylatio n (Fig. 8C). Phosphorylation at Y972 Does Not Affect SH2BMediated Y1007/Y1008 Phosphorylation. SH2Bis another important component of Jak2 si gnal transduction. It has been shown to be a potent Jak2 activator (22). We sought to determine whether Y972 phosphorylation affects SH2Bmediated Jak2 activation. COS-7 cells were transiently transfec ted with either Jak2 wild-type or Jak2 Y972F plasmid. Additionally, th ese cells were co-transfected with increasing amounts of a myc-tagged SH2Bconstruct. All cells were lysed and Jak2 protein was immunoprecipitated from the lysates. Ja k2 Y1007/Y1008 phosphorylation was measured via western blot. The loss of Y972 phos phorylation did not hinder SH2Bmediated Y1007/Y1008 phosphorylation (Fig. 9A). The membrane was stripped and re-probed with an anti-Jak2 polyclonal antibody to verify equal protein load ing (Fig. 9B). The expression of myc-tagged SH2Bprotein was also verified vi a western blot (Fig. 9C). The Loss of Y972 Phosphorylation Impairs Growth Hormone-Mediated STAT1 Nuclear Translocation. In the previous signaling experiments, we i nvestigated the effect of Y972 phosphorylation on the response of Jak2 to several different activators. We then decided to look at the downstream consequences of losing Y972 phosphorylation. 2A cells that stably express the growth hormone receptor we re transfected with 10 g of a plasmid encoding a GFP-tagged STAT1 protein. Additionally, these cells were co-transfected with 10 g of either Jak2 wild-type

PAGE 48

48 plasmid or Jak2 Y972F plasmid. The cells were treated with growth hormone (250 ng/ml) for 0 and 30 minutes. The cells were then viewed under a fluorescent microscope to determine the position (nucleus vs. cytoplasm) of the GFP-ST AT1 protein. The loss of Y972 phosphorylation impaired GFP-STAT1 nuclear tran slocation, as compared to Jak2 wild-type protein (Fig. 10A). The cells were stained with DAPI to identify the nucleus (Fig. 10B). The two images were then merged (Fig. 10C). The GFP fluorescence intensity was quantified using Image J software. The nuclear/cytoplasmic intensity ratio was calcula ted and taken as an indicator of GFP-STAT1 nuclear translocation. The cells transfected with Jak2 Y972F mutant plasmid showed lower ratios at both time points (Fig. 10D). Discussion One of the most notable findings of th is study is that, even though the loss of phosphorylation at Y972 produced dramatic reduc tions in Jak2 tyrosine phosphorylation, the Jak2 Y972F protein was still able to drive gene e xpression of the luciferase reporter construct. Given the significant loss of phosphorylation at Y1007 in response to the Y972F mutation, we expected to see a similarly drastic reducti on in functional output. However, the Y972Fdependent loss in Y1007 phosphorylation did not affect Jak2-mediated luciferase gene expression. A possible explanation for this di sparity is that while Y1007 phosphorylation is thought to be necessary for maximal Jak2 activat ion, Jak2 can mediate gene expression in the absence of Y1007 phosphorylation (85) It is entirely possible that, within the confines of the luciferase assay, only the basal le vel of Jak2 dependent gene expr ession can be measured. Thus, Y972 phosphorylation may only affect maximal Jak2 activation and suboptim al activation states may be left intact. Tyrosine 972 was also investig ated within a signaling cont ext, and these results have significant biochemical implications. In this study, we used the angiotensin II and growth

PAGE 49

49 hormone signaling pathways as examples of the two Jak2-dependent signaling paradigms. One of the key differences between th ese two models is that while cy tokine receptors are thought to activate Jak2 molecules by bringing them into close physical contact with one another, no evidence exists to suggest that GP CRs facilitate such intimate cont act. Thus, the fact that Y972 phosphorylation is important for angiotensin II dependent Jak2 to tal phosphorylation and independent of growth hormone dependent Jak2 total phosphorylation may be a result of this difference. It is possible that the loss of Y972 phosphorylation im parts a degree of instability to Jak2 dimers, which can be overcome with the aid of a strong dimerizing agent like a cytokine receptor. The ligand independent losses in bo th Jak2 total and Y1007 phosphorylation seen in response to the Y972F mutation are also in ag reement with this model since no dimerizing agents were used in these experiments. Further support for this m odel arises when comparing the effects of losing Y972 phosphorylation on growth hormone-depe ndent Y1007 phosphorylation and SH2Bmediated Y1007 phosphorylation. Both the grow th hormone receptor and SH2Bfacilitate Jak2-Jak2 interaction. However, SH2Bis thought to be an enhancer of growth hormone dependent Jak2 activation, which suggests that it ad ds an additional increment of st ability to Jak2 dimers that the growth hormone receptor is unable to provide. Th e putative stability deficit produced by the loss of Y972 phosphorylation may be too great to be fully overcome by growth hormone receptor activation. This would explain why growth hor mone receptor activation largely restored Jak2 global phosphorylation, but not Y1007 phosphorylat ion, in the Y972F mutant. Maximal activation, as indicated by Y1007 phosphorylatio n, was achieved by the presence of SH2B. This suggests that SH2Bwas able to provide an extra degree of stability to the mutant, allowing it to become maximally activated.

PAGE 50

50 Finally, this model can be extended to explain the fact th at the loss of Y972 phosphorylation reduced the ability of Jak2 to mediate growth hormone-dependent STAT1 nuclear translocation. It is possi ble that the suboptimal activati on state of the Y972F mutant may have been insufficient to achieve full mediation of STAT1 nuclear translocation. The fact that the Y972F mutant is capable of activating STAT 1 to the same extent as wild-type Jak2 in a ligand independent manner does not contradict th is hypothesis because it has been previously shown that STAT1 phosphorylation is not the sole factor behi nd STAT1 nuclear translocation (86). Other elements, such as STAT1 recruitment to the growth hormone receptor may have played a role in this reducti on. The Y972F mutant may have l acked sufficient kinase power to accomplish all the tasks associated with STAT 1 nuclear translocation while retaining enough kinase function to fully activate STAT1. The f act that the loss of Y 972 phosphorylation did not eradicate STAT1 nuclear translocation bolsters th is assertion. Thus, th e results of the STAT1 nuclear translocation experiment expand the mode l of a Y972-dependent loss of stability in Jak2 dimerization by showing that this putative phenomenon has consequences for downstream elements of Jak2-dependen t signal transduction. In conclusion, the loss of phosphorylation at Y972 has significant consequences for Jak2 biochemistry and signal transduction. We assert that this loss c onfers a degree of instability to the maximally active Jak2 conformation. This in stability hinders Jak2 kinase function, as is demonstrated by the fact that the Y972F muta nt could not fully phosphorylate HA-tagged wildtype Jak2. It also manifests itself in the seve re reductions in both Y1007 and Jak2 total tyrosine phosphorylation. With respect to signal tran sduction, the putative Y 972-dependent stability deficit is a hurdle to activating agents that have not been show n to facilitate Jak2 dimerization, such as the AT1 receptor. Finally, this lack of stab ility may hinder the full propagation of a

PAGE 51

51 Jak/STAT signal, as was seen in the growth hor mone-dependent STAT1 nucl ear translocation. It is also important to note that the loss of Y972 phosphorylation does not irreversibly hinder Jak2 function. A potent activator such as SH2Bwas able to restore maximal activation to the Y972F mutant. Furthermore, suboptimal levels of activation seem to be unaffected by the loss of Y972 phosphorylation. Thus, we asse rt that Y972 phosphorylation is necessary for maximal Jak2 function.

PAGE 52

52 Figure 3-1. Tyrosine 972 is a site of Jak2 aut ophosphorylation. Mass spectrometry analysis was performed on purified Jak2 protein. It was determined that Y972 is a Jak2 phosphotyrosine. As controls, the phosphor ylation statuses of Y221 and Y1007, two known phosphotyrosines (19,22), were verified as well. This data was generated by Dr. Xianyue Ma.

PAGE 53

53 Figure 3-2. Tyrosine 972 is solvent accessible. Using the Swiss Model program, we generated a structural homology model of murine Jak2 based on the human Jak2 crystal structure (pdb code 2B7A). We then used the Defi nition of Secondary Structure of Proteins (DSSP) program (80) to calculate the solv ent accessibility of Y972. The solventaccessible area of Y972 is 26 2. Tyrosine 972 is highlighted in red.

PAGE 54

54 Jak2 (pY1007/Y1008)IP: Jak2 pAb IB: anti-Jak2 pY1007/Y1008 pAb IB: anti-Jak2 pAb 111 111 Total Jak2B.IP: Anti-Jak2 pAb pRC-CMV Empty Vector pRC-CMV Jak2 Y972FA. pRC-CMV Jak2-WT + + + IB: Anti-Tyr(P)-mAb IB: Anti-Jak2 pAb 111 111C. D. Figure 3-3. The effect of Y972 phosphorylation on Ja k2 total and Y1007 phosphorylation. A). BSC-40 cells were transfected with 5 g of the indicated plasmid DNA. These constructs were overexpressed using vaccini a virus-delivered T7 RNA polymerase. Western blot analysis shows that the loss of phosphorylation at Y972 drastically reduces Jak2 total tyrosine phosphorylation. B). The membrane was stripped and reblotted with an anti-Jak2 polyclonal antibody to verify equal protein loading. C). COS-7 cells were transfected with 10 g of either pBOS-Jak2 WT or pBOS-Jak2 Y972F. Immunoblotting with an anti-pY1007/ Y1008 pAb revealed that the loss of phosphorylation at Y972 significantly re duced Y1007 phosphorylation. D). The membrane was stripped and re-blotted with an anti-Jak2 polyclonal antibody to verify equal protein loading.

PAGE 55

55 pCI HA Jak2 WT ( g) 5 5 5 5 5 5 5 15 10 10 10 5 15 5 15 pRC-CMV-Jak2 WT ( g) pCI Jak2 DN ( g) pRC-CMV-Jak2 Y972F ( g) IP: anti-HA mAb IB: anti-Jak2 pY1007/Y1008 pAbIB: anti-HA pAb Jak2 (pY1007/Y1008) HA-Jak2 111 111A.pB/S-Empty Vector ( g) 5 15B.Jak2 (pY1007/Y1008)pBOS-Jak2 WT ( g) pCI-HA Jak2 Y972F ( g) pCI-HA Jak2 WT ( g) pRC-CMV E.V. ( g)C.IP: anti-HA mAb111 111IB: anti-HA pAb Total Jak2 1 3 10 1 3 10 10 10 10 10 10 10 IB: anti-Jak2 pY1007/Y1008 pAb 9 7 9 7D. pCI-HA-Jak2-WT ( g) pRK5-FLAG-Jak2 Y972F ( g) pRK5-FLAG-Jak2 WT ( g) pRC-CMV-Empty Vector ( g) 10 10 10 10 10 10 IP: anti-HA mAb IB: anti-FLAG pAb IB: anti-HA pAb FLAG-Jak2 HA-Jak2 111 111E. F. Figure 3-4. Tyrosine 972 phosphor ylation affects Jak2 kinase f unction, but not its substrate properties or Jak2 dimerization. A). The indicated plasmids were overexpressed in BSC-40 cells and Y1007/Y1008 phosphorylation was determined via western blot. The loss of Y972 phosphorylation did not produ ce an inhibitory phenotype but it did impair Jak2 kinase function, as compared to wild type protein. B). The membrane was stripped and re-probed to verify e qual loading. C). The effect of Y972 phosphorylation on Jak2 substrate potentia l was also evaluated. The indicated plasmids were overexpressed in BSC40 cells and the Y1007/Y1008 phosphorylation of HA-Jak2 wild-type protein was measured via western blot. The loss of Y972 phosphorylation did not prevent Jak2 from being phosphorylated at the Y1007/Y1008 positions. D). Equal protein loading was verifi ed via western blot. E). The effect of Y972 phosphorylation on Jak2 dimerization wa s evaluated. The indicated plasmids were expressed in COS-7 cells a nd FLAG Jak2/HA-Jak2 coprecipitation was measured via western blot. Dimerizati on was shown to be independent of Y972 phosphorylation. F). The membrane was st ripped and reblotted with an anti-HA polyclonal antibody to determine protein loading.

PAGE 56

56 pBOS-Jak2 WT ( g) pBOS-Jak2 Y972F ( g)1 3 10 1 3 10 IP: anti-STAT1 pAb IB: anti-STAT1 pY701 pAb IB: anti-STAT1 pAb STAT1 (P) STAT1 79 79A. B. Figure 3-5. The loss of tyrosine 972 phosphoryl ation does not affect Jak2-mediated STAT1 activation. COS-7 cells were transfected with in creasing amounts of either Jak2 wildtype or Jak2 Y972F plasmid. Lysates were prepared and STAT1 protein was immunoprecipitated from the lysates. A) Phosphorylation at Y701 in STAT1 was verified via western blot. The loss of Y972 phosphorylation did not hinder Jak2mediated STAT1 activation. B). Equal STAT 1 protein loading was also verified via western blot.

PAGE 57

57 Jak2-Dependent Luciferase Gene Expression0 1 2 3 4 5 6 1310 Jak2 Plasmid (micrograms)RLU Fold Change Jak2 WT Jak2 Y972F A. Jak2 WTJak2 Y972F NT Whole Cell Lysates IB: anti-Jak2-pAbB. Figure 3-6. Tyrosine 972 does not affect ligandindependent gene expres sion. COS-7 cells were transfected with 2 g of a firefly luciferase plasmid containing four tandem repeats of the activating sequence. Additionally, th ese cells were co-transfected with increasing amounts (1, 3, or 10 g) of either Jak2 wild-type or Jak2 Y972F plasmids. A). The cells were trypsinized and seeded into six-well plates. The cells were lysed and a 20 l sample of each lysate was treated with 100 l of luciferase substrate. Relative light units were read by a lumino meter and were taken as a measure of luciferase gene expression. The loss of Y972 phosphorylation did not hinder Jak2dependent, luciferase gene e xpression. B). A portion of the trypsinized cells were used to verify equal protein expression of Jak2 wild-type and Jak2 Y972F proteins. Equal expression was determined via we stern blotting with an anti-Jak2 pAb.

PAGE 58

58 Figure 3-7. Tyrosine 972 is critical for angiotensin II-dependent Jak2 phosphorylation. COS-7 cells were transiently transfected with AT1 receptor plasmid and either empty vector plasmid, Jak2 wild-type plasmid, or Jak2 Y 972F plasmid. After transfection, these cells were treated with 100 nM angiotensi n II for 0, 3, and 6 minutes. The cells were lysed and Jak2 protein was immunoprecipitated from the lysates. A). Jak2 tyrosine phosphorylation was determined via wester n blot. The loss of Y972 phosphorylation severely reduced angiotensin II-dependent Jak2 tyrosine phosphorylation. B). The membrane was stripped and re-probed with an anti-Jak2 pAb to verify equal protein loading.

PAGE 59

59 GH (min) 0 5 10 0 5 10 0 5 10 Jak2 (pY1007/1008) 111 111 IB: anti-Jak2-pAb Jak2 Whole Cell Lysate IB: anti-Jak2-pY1007/1008-pAb Empty VectorJak2 WTJak2 Y972FA. B. Jak2-WT Jak2-Y972F Empty Vector 111 Jak2-(P) GH (min) 0 5 10 0 5 10 0 5 10 IP: anti-Tyr (P)-mAb IB: anti-Jak2 pAbC. Figure 3-8. Tyrosine 972 differentially affects growth hormone-dependent Jak2 total phosphorylation and Y1007 phosphorylation. In two separate experiments, COS-7 cells were transiently transfected with growth hormone receptor plasmid and either empty vector plasmid, Jak2 wild-type pl asmid, or Jak2 Y972F plasmid. A). Phosphorylation at Y1007/Y1008 was measur ed via western blot. The loss of phosphorylation at Y972 significantly re duced growth hormone dependent Y1007/Y1008 phosphorylation. B). Equal pr otein loading was determined via western blot. C). Jak2 total phosphorylati on was also measured via western blot. Tyrosine 972 did not affect growth-hormone dependent increases in Jak2 total phosphorylation.

PAGE 60

60 IP: Jak2 pAb Jak2 WT Jak2 Y972F Jak2 (pY1007/Y1008) Total Jak2 111111 IB: anti-Jak2 pY1007/Y1008 pAb IB: anti-Jak2 pAbA. B.79Whole Cell Lysates myc-SH2BIB: anti-myc-mAbC. Figure 3-9. Jak2 can be activated by SH2Bin the absence of tyrosine 972 phosphorylation. A). COS-7 cells were transfected with increasing amounts (0, 0.5 and 1 g) of a myc-tagged SH2Bplasmid and either Jak2 wild-type or Jak2 Y972F plasmid. Phosphorylation at Y1007/Y1008 was measur ed via western blot. The loss of phosphorylation at Y972 did not affect SH2Bmediated Y1007/Y1008 phosphorylation. B). The membrane was st ripped and re-probed to verify equal protein loading. C). The levels of myc-SH2Bprotein were determined via a western blot of whole cell lysate samples.

PAGE 61

61 MERGE Jak2-WT Jak2-Y972F 0 0 30 30 GFP DAPI GH (min ) 0 0 0.5 1 1.5 2 2.5 3 030 growth hormone (min) GFP-STAT1 Fluorescence (Nucleus/Cytoplasm) Jak2 WT Jak2 Y972F A.B. C. D. Figure 3-10. Tyrosine 972 phosphorylation a ffects GFP-STAT1 nuclear translocation. 2A cells that stably express the growth hormone r eceptor were transiently transfected with a GFP-STAT1 plasmid and either Jak2 wild -type plasmid or Jak2 Y972F plasmid. These cells were treated with 250 ng/ul gr owth hormone for 0 and 30 minutes. The location of the GFP-STAT1 molecule was determined via fluorescence microscopy. A). The loss of Y972 phosphorylation re duced growth hormone-dependent nuclear translocation of the GFP-STAT1 protein. B). The cells were subjected to DAPI staining to indentify the cell nucleus. C). The two images were then merged. D). The GFP fluorescence intensity was qua ntified using Image J software. Nuclear/cytoplasmic intensity ratios were calculated and graphe d as a function of time. The cells transf ected with Jak2 Y972F plasmid showed lower nuclear/cytoplasmic intensities at both time points.

PAGE 62

62 CHAPTER 4 S1120 AND JAK2 FUNCTION1 Summary Janus Kinase 2 (Jak2) is an integral component of the grow th hormone signaling pathway. The Jak/STAT pathway allows the growth hormone receptor to alter gene transcription patterns in the nucleus, thereby promoting growth and differe ntiation. In order to maintain this nuclear link, Jak2 function must be optimal. In this study we have identified and characterized a serine residue that is critical for Jak2 function. Serine 1120 is located in the Jak2 kinase domain. The mutation of this residue to alanine prevente d Jak2 from achieving autophosphorylation in a ligand independent system. It al so severely hindered ligand inde pendent gene expression. In a ligand-dependent context, several elements of growth hormone-dependent signal transduction were affected by the conversion of S1120 to alan ine, including growth hormone dependent Jak2 activation, SH2B-beta mediated Ja k2 activation, and growth hormone dependent luciferase gene expression. Thus, we assert that this residue is an essential component of proper Jak2 function. Introduction Growth hormone is a peptide horm one released from the anterior pituitary (87) It acts on multiple target tissues to promote growth and differentiation (87). This hormone can easily become a double-edged sword as it has been implicated in both physiological and pathophysiological processes. Adequate levels of this hormone are necessary during human development in order to prevent a short stature (8 7). In contrast, an excess of growth hormone can lead to gigantism and cancer (87). Thus, exquisite regulation of growth hormone is critical for normal human health. 1 Reprinted with permission from Dr. David Ostrov, Dr. Peter P. Sayeski, and Andrew Magis.

PAGE 63

63 The growth hormone signal is transduced by a number of different intracellular signaling pathways, including the Jak/STAT pathway. Janus kinase 2 (Jak2) is a member of the Janus family of tyrosine kinases and transduces si gnals from cytokine receptors, like the growth hormone receptor. Jak2 molecules are constituti vely bound to the cytoplasmic tail of the growth hormone receptor. The role of the growth hormo ne receptor is to translate a growth hormone binding event into a series of conformationa l changes that place the constitutively bound Jak2 molecules in close enough proximity to one another to achieve autophosphorylation. The phosphorylated Jak2 molecules then phosphorylate th e receptor tail, facili tating the recruitment of STAT proteins to the receptor. The recr uited STAT proteins ar e phosphorylated by Jak2, which leads to their dimerization. The dimerized STATs then translocate to the nucleus and bind DNA promoter elements to alter gene transcrip tion patterns. Thus, Jak2-dependent signal transduction provide s growth hormone with an important link to the cell nucleus. With such a link in place, growth hormone can alte r the expression of genes that are important for growth and differentiation. To maintain this link, it is absolutely critical that Jak2 is able to function in an optimal manner. Since the initial cloning of Jak2 in the early 1990s (1), much work has been done in an effo rt to understand the f unction of this signaling intermediate. It is clear that a very tight rela tionship exists between the structure and function of Jak2. Several amino acids have been as being important in maintaining proper Jak2 function. Many of these residues are phosphotyrosines th at have been linked to key regulatory mechanisms of Jak2 function (19-24). Other non -tyrosine residues have been characterized in the Jak2 kinase domain, and have been shown to be essential for maintaining proper kinase function (14-17).

PAGE 64

64 In 2006, the first report of a site of serine phosphorylati on in Jak2 was published (88). This report represented a major development in the Jak2 field because prior work had only implicated tyrosines in phosphorylation-dependent regulation of Jak2. Se rine 523 resides in the Jak2 pseudokinase domain and has been shown to be phosphorylated in response to growth hormone treatment (88). Its phos phorylation is thought to repr ess growth hormone-dependent Jak2 function as its mutation to alanine enhanced growth hormone dependent Jak2 activation and STAT5 phosphorylation (88). Here, we have identified another serine resi due that signif icantly affects Jak2 function. Serine 1120 is located in the Jak2 kinase domain. Its mutation to alanine is deleterious for Jak2 kinase function. Several aspects of Jak2-de pendent signaling, such as growth hormonedependent Jak2 activation, SH2Bmediated Jak2 activation and growth hormone-mediated gene transcription were shown to be S1120-depende nt. We assert that S1120 is a novel site of Jak2 functional regulation. Results Serine 1120 is Solvent Accessible We performed in silico modeling of the Ja k2 kinase domain to determine the solventaccessible surface area of S1120. Using the Sw iss model program, a structure homology model of the murine Jak2 kinase domain was generated based on the crystal structure of the human Jak2 protein (protein database code: 2B7A). Using th e Definition of Secondary Structure of Proteins (DSSP) program (80), we determined that the solvent accessible surface area of S1120 is 26 2. As 15 2 is generally considered the minimum su rface area required for a phosphorylation site, the solvent accessible surface area of S1120 is co nducive to phosphorylation. We have shown a representation of two dimerized Jak2 molecules with S1120 highlighted in green. Based on this model, S1120 is thought to be in close proxi mity to the Jak2 dimer interface (Fig. 1).

PAGE 65

65 Serine 1120 is Critical for Jak2 AutophosphorylationJak2 tyrosine autophosphorylation is a cri tical precursor to Ja k2-dependent signaling events such as receptor and ST AT protein phosphorylation. Theref ore, we decided to evaluate the ability of the Jak2 S1120A mutant to ach ieve autophosphorylation via overexpression. BSC40 cells were transiently transfected with 5 g of either empty vector plasmid, Jak2 wild-type plasmid, or Jak2 S1120A plasmid. Using the vaccinia virus overexpr ession system, these plasmids were overexpressed in BSC-40 cells. Si xteen hours after vaccinia virus infection, all cells were lysed and Jak2 protein was immunopr ecipitated from the lysates. Jak2 tyrosine phosphorylation levels were determined via west ern blot. The mutation of S1120 to alanine abolished the ability of Jak2 to achieve autophosphor ylation (Fig. 2A). The membrane was stripped and re-probed to verify protein loading (Fig. 2B). The S1120A Mutant Displays a Mild Inhibitory Phenotype We evaluated the ability of the Jak2 S1120 mutant to act as a dominant negative molecule. COS-7 cells were transiently transfected with 5 g of an HA-tagged Jak2 wild-type plasmid. Additionally, these cells were tr ansfected with increas ing amounts (5 g and 15 g) of either a Jak2 wild-type plasmid, a Jak2 dominant negative plasmid, or the Jak2 S1120A plasmid. Thirtysix hours after transfection, all cells we re lysed and HA-tagged Jak2 protein was immunoprecipitated from the lysates. The ty rosine phosphorylation of HA-tagged Jak2 wildtype protein was measured via western blot. While titration of the Jak2 wild-type plasmid increased HA-tagged Jak2 wild-type phosphorylati on, increasing amounts of either the dominant negative or Jak2 S1120A plasmid did not in crease HA-tagged Jak2 wild-type phosphorylation (Fig. 3A). In fact, the Jak2 S1120A mutant slightly decreased HA-tagged Jak2 wild-type phosphorylation (Fig. 3A). The membrane was stripped and re-probed with an anti-Jak2 polyclonal antibody to verify equa l protein loading (Fig. 3B).

PAGE 66

66 The Jak2 S1120A Mutation Hinders Liga nd-Independent Gene TranscriptionLigand-independent experiments like the BS C-40 autophosphorylation assay (Fig. 2) evaluate the intrinsic abilities of Jak2. In addition to achieving ligand-independent phosphorylation, Jak2 can also medi ate gene transcription. Thus we looked at how the S1120A mutation affected the intrinsic ability of Jak2 to drive gene transcrip tion. COS-7 cells were transiently transfected with 2 g of a lucifera se plasmid under the control of a STAT-responsive promoter. Additionally, these ce lls were transfected with in creasing amounts of either Jak2 wild-type plasmid or Jak2 S1120A plasmid. These cel ls were then trypsinized and seeded into six well plates. A portion of the trypsinized cells were seeded into 100 mm dishes to be used for verification of equal Ja k2 protein expression. Forty-eight hours after transfect ion, all cells were lysed for five hours and exposed to one freeze-thaw cycle at -80C. A 20 l sample of each lysate was then exposed to an excess (100 l) of luciferase substrate. Luminescence wa s read by a luminometer and was taken as an indicator of luciferase gene expression. Luciferase gene experession was compared between Jak2 wild type and Jak2 S1120 at the 3X and 10X dosage points using a two-tailed T Test. It was determined that the S1120A mutation seve rely hindered the abil ity of Jak2 to drive luciferase gene expression (Fig. 4A). Equal Jak2 protein expression wa s verified via western blot (Fig. 4B). Growth Hormone-Dependent Jak2 Activation is Dependent on S1120 Jak2 is an integral component of the growth hormone signaling pathway. In response to growth hormone treatment, Jak2 becomes activated. As this is a necessary prerequisite for all subsequent Jak2-dependent signali ng events in the growth hormone pathway, we looked at the effect of the S1120A mutation on Jak2 activati on in response to growth hormone treatment. COS-7 cells were transiently transfected with 10 g of growth hormone receptor plasmid.

PAGE 67

67 Additionally, these cells were transfected with 3g of Jak2 wild-type pl asmid and 10 g of Jak2 S1120A plasmid in order to maintain equal expression levels. Twenty-four hours after transfection, the cells we re serum-starved for 16 hours. The cel ls were then treated with growth hormone (250 ng/ml) for 0, 5 and 10 minutes. All cells were lysed and Jak2 protein was immunoprecipitated from the lysates. The phos phorylation status of Y1007/Y1008 in the Jak2 activation loop was determined via west ern blot. While Y1007/Y1008 phosphorylation increased in wild-type Jak2 in a time-dependent manner, there was no such increase in the S1120 mutant (Fig. 5A). The membrane was stripped and re-probed to verify equal protein loading (Fig. 5B). Serine 1120 is Critical for SH2BMediated Jak2 ActivationSH2Bbinds Jak2 at Y813 and is thought to pr omote Jak2 activation by stabilizing the active Jak2 conformation (21,35). This protein is an important component of the growth hormone signaling pathway as it has been show n to significantly enhance growth hormone dependent Jak2 activation. We sought to determ ine the effect of the S1120A mutation on SH2Bmediated Jak2 activation. COS-7 cells were tr ansiently transfected with increasing amounts of myc-tagged SH2Bplasmid and fixed amounts of either Jak2 wild-type plasmid or Jak2 S1120A plasmid. Thirty-six hours after transfec tion, all cells were lysed and Jak2 protein was immunoprecipitated from the lysates. Th e phosphorylation status of Y1007/Y1008 was determined via western blot. The S1120A mutation eliminated the ability of Jak2 to be activated by SH2Bwhen compared to Jak2 wild-type control. (Fig. 6A). The membrane was stripped and re-probed to verify equal protein load ing (Fig. 6B). The expression of myc-SH2Bprotein was verified by western blotting whole cell ly sate samples with an anti-myc monoclonal antibody (Fig. 6C).

PAGE 68

68 The S1120A Mutation Abolishes Growth Hormone-Mediated Ge ne Transcription The ultimate cellular response to the activation of Jak/STAT signaling is the alteration of gene transcription patterns. We sought to de termine whether S1120 is critical for growth hormone-mediated changes in gene transcription. COS-7 cells were transiently transfected with 2 g of luciferase plasmid under the control of a STAT-inducible promoter. Additionally, these cells were transfected with 10 g of a growth hormone receptor and either an empty vector plasmid, Jak2 wild-type plasmid, or a Jak2 S1120 plasmid. These cells were trypsinized and seeded into six-well plates at 2 x 105 cells per well. A portion of th e trypsinized cells were also seeded into 100 mm dishes to be used for verification of equal Jak2 protein expression. Forty-eight hours after transfect ion, all cells were serum starved for 12 hours. The cells were treated with growth horm one for 0, 4, 8, 12 and 24 hours. Th e cells were then lysed and a 20 l sample of each lysate was treated with an excess (100 l) of luciferase substrate. Luminescence was measured with a luminometer a nd was taken as an indicator of luciferase gene expression. Cells transfected with Jak2 w ild-type protein produced luminescence readings that increased over the 24 hour time course (Fi g. 7A). In contrast, cells expressing the Jak2 S1120A mutant did not produce luminescence read ings above those tran sfected with empty vector plasmid (Fig. 7A). Equal Jak2 protein expression wa s verified via wester n blot (Fig. 7B). Discussion Jak2 plays a critical role in growth horm one signaling. The Jak/STAT pathway provides the growth hormone receptor with an important link to the cell nucleus. Thus, it is essential that Jak2 functions optimally in order to maintain this nuclear link. In this study, we have identified S1120 as an amino acid residue that is critical for Jak2 function. We have shown that this residue is essential for Jak2 kinase function as its mutation to al anine eliminated the autophosphorylative capacity of Ja k2 and hindered its ability to drive ligand-independent gene

PAGE 69

69 transcription. We have also demonstrated that this loss of function ha s several consequences, especially for growth hormone signaling. Sp ecifically, the S1120-depe ndent loss of function affects Jak2 activation by both growth hormone and SH2B. Finally, we demonstrated that this residue is critical for growth hor mone-dependent gene transcription. In this study, we utilized s ite-directed mutagenesis to mutate S1120 to alanine. This mutation is a relatively conservative change th at involves the loss of the OH functional group on S1120. An in silico analysis of the putative interac tions between S1120 and other Jak2 residues suggests that this loss does not affect any existing intramolecular interactions. Based on this data and other observations, we have iden tified three possible mechanistic explanations for the S1120A effect on Jak2 function; 1). Serine 11 20 may be a site of serine phosphorylation in murine Jak2, 2). The S1120 OH functional group ma y facilitate interaction with a putative Jak2 binding partner, 3). The S1120A mutation may conf er an inhibitory phenotype, which severely hinders Jak2 function. We will explore each of these hypotheses individually, but it is entirely possible that the S1120A effect may be the re sult of a synergy of these three elements. In 2006, the first site of serine phosphorylation in Jak2 was id entified as S523 (88). This site was shown to be phosphorylated in res ponse to growth hormone treatment and was characterized as a repressor of growth horm one-dependent Jak2 signaling as its mutation to alanine enhanced this process (88). The idea th at S1120 is a site of serine phosphorylation is supported by the fact that it is solvent accessible. However, S1120 phosphorylation would putatively play a much different role from that of S523 phosphorylation as its mutation to alanine eliminated growth hormone-dependent Jak2 si gnaling. Thus, in contrast to S523, S1120 phosphorylation would potentially enha nce growth hormone signaling.

PAGE 70

70 As previously mentioned, the mutation of S1120 to alanine involve s the loss of an OH functional group. While this change can certainly affect the poten tial phosphorylation status of S1120, it may have other consequences as well. The solvent accessibility of S1120 raises the possibility that it facilitates interaction with a Jak2 binding partner. Th is putative interaction may promote Jak2 stability and its loss may explain why a signi ficantly greater amount of Jak2 S1120A plasmid must be transfected in comparison to Jak2 wild-type plasmid in order to achieve comparable expression levels. Finally, the S1120A mutation may confer an inhi bitory phenotype to Jak2 that interferes with its function. While preliminary, the data in Figure 3 suggest that th e conversion of S1120 to alanine produces a mild domina nt negative phenotype. This i nhibitory character may explain why both ligand dependent and ligand indepe ndent gene expression was hindered by the conversion of S1120 to alanine. Future experi ments should evaluate the utility of the Jak2 S1120A mutant as method of Jak2 inhibition. In conclusion, the data in this study suggest that S1120 is critical for murine Jak2 function. The mutation of S1120 to alanine had profound e ffects on Jak2 function, both on an intrinsic level and within a signaling context. The impor tance of S1120 to murine Jak2 function may be the result of several differe nt factors. First, S1120 may be a site of serine phosphorylation within murine Jak2. The precedent exists for th e serine phosphorylation of Jak2 and it has consequences for Jak2 function (88) Serine 1120 may also be cri tical for Jak2 interactions with other proteins. Finally, this mutation may confer an inhibitory phenotype to Jak2. While these factors are certainly independent from one another, the S1120A effect may result from a combination of the three elements. The phosphor ylation of S1120 could potentially provide a protein binding site. Finally, a dominant nega tive phenotype may result from the loss of the

PAGE 71

71 interaction with this putative Ja k2 binding protein. Future studies should aim to mechanistically elucidate the role of S1120 in Jak2 function.

PAGE 72

72 Figure 4-1. Serine 1120 is solven t accessible. Using the Swiss Model program, we generated a structural homology model of murine Jak2 based on the human Jak2 crystal structure (pdb code 2B7A). Using the COOT program we superimposed the murine model on human Jak2. We then used the Definition of Secondary Structure of Proteins (DSSP) program (80) to calculate the solvent accessibility of S1120. The solvent-accessible area of S1120 is 26 2. Serine 1120 is highlighted in green at the dimer interface.

PAGE 73

73 IP: anti-Jak2 pAb IB: anti-P.Tyr mAb IB: anti-Jak2 pAb 111 111A. B. Figure 4-2. The importance of S1120 for Ja k2 autophosphorylation. BSC-40 cells were transfected with 5 ug of the indicated plasmids. These constructs were overexpressed using a vaccinia virus overexpression system. A). Western blot analysis reveals that the S1120A mutation prevents Jak2 autophos phorylation. B). The membrane was stripped and re-probed with an anti-Jak2 polyclonal antibody to verify equal protein loading.

PAGE 74

74 Figure 4-3. The Jak2 S1120A mutant displays a m ild inhibitory phenotype. COS-7 cells were transiently transfected with HA-tagged Jak2 wild-type plasmid and increasing amounts of either a Jak2 w ild-type plasmid, a Jak2 dominant negative plasmid, or the Jak2 S1120A plasmid. All cells were ly sed and HA-tagged Jak2 wild-type protein was immunoprecipitated from the lysates. A) Western blot analysis revealed that the Jak2 S1120A mutant mild ly inhibited HA-tagged Ja k2 wild-type phosphorylation. B). The membrane was stripped and re-pro bed with an anti-Ja k2 polyclonal antibody to verify equal protein loading.

PAGE 75

75 Figure 4-4. Serine 1120 is important for ligand-independent gene transcription. A). COS-7 cells were transfected with 2 ug of luciferase plasmid and increasing amounts of either Jak2 wild-type plasmid or Jak2 S1120A plasmi d. Empty vector plasmid was used to ensure that all cells were transfected with equal amounts of plasmid. A luciferase gene reporter assay reveal ed that the Jak2 S1120A muta nt could not achieve the levels of luciferase gene e xpression that the wild-type pr otein could. B.) A portion of the transfected cells were used for wester n blotting in order to verify equal Jak2 protein expression. These cells were lysed a nd a sample of each lysate was subjected to SDS-PAGE. The proteins were then tr ansferred to a nitrocellulose membrane. Western blot analysis with an anti-J ak2 polyclonal antibody revealed that comparable Jak2 protein expression wa s achieved at the 3X dosage point.

PAGE 76

76 IP: anti-Jak2 pAb GH (min) 111 111 IB: anti-pY1007/Y1008 pAb IB: anti-Jak2 pAb Jak2 WT Jak2 S1120A 0 5 100 5 10 A. B Figure 4-5. Growth hormone-dependent Jak2 activ ation is eliminated by the S1120A mutation. COS-7 cells were transiently transfected w ith the indicated plasmids. Twenty-four hours after transfection, all ce lls were serum-starved for 16 hours. The cells were then treated with growth hormone (250 ng/ ml) for 0, 5, and 10 minutes. All cells were then lysed and Y1007/Y1008 phosphorylat ion status was assayed via western blot. A). It was determined that the S1120A mutation prevented the growth hormone-dependent increase in Y1007/Y 1008 phosphorylation. B). The membrane was stripped and re-probed with an an ti-Jak2 polyclonal antibody to verify equal protein loading.

PAGE 77

77 Jak2 WTJak2 S1120A IP: anti-Jak2 pAb IB: anti-pY1007/Y1008 pAb IB: anti-Jak2 pAb 111 111 Whole Cell Lysate IB: anti-myc mAb myc-SH2B0 0.5 10 0.5 1A. B. C. 79 Figure 4-6. Serine 1120 is criti cal for SH2B-beta mediated Jak2 activation. COS-7 cells were transiently transfected with increasing amounts of myc-SH2Bplasmid and either Jak2 wild-type plasmid or Jak2 S1120A plas mid. Thirty six hour s after transfection, all cells were lysed and Jak2 protein was i mmunoprecipitated from the lysates. A). The phosphorylation status of Y1007/Y1008 wa s determined via western blot. The conversion of S1120 to al anine eliminated SH2Bdependent increases in Y1007/Y1008 phosphorylation. B). The membra ne was stripped and re-probed with an anti-Jak2 polyclonal antibody to verify equal protein loading. C). Additionally, whole cell lysate samples were subjected to SDS-PAGE and transferred to a nitrocellulose membrane. This membrane was blotted with an anti-myc monoclonal antibody to determine equal levels of SH2Bexpression.

PAGE 78

78 Figure 4-7. Serine 1120 is importa nt for growth hormone dependent gene transcription. COS-7 cells were transiently transfected with a growth hormone receptor plasmid and a luciferase plasmid. Additionally, these cells were transfected with either an empty vector plasmid, a Jak2 wild-type plasmid or a Jak2 S1120A plasmid. All cells were then trypsinized and seeded into six well pl ates. A). A luciferase gene reporter assay revealed that the Jak2 S1120A mutant did not mediate luciferase gene transcription at levels significantly above t hose seen for the empty vector plasmid. B). A portion of the trypsinized cells were seeded into 100 mm dishes and used to verify the expression levels of Jak2 wild-type prot ein and Jak2 S1120A prot ein. Western blot analysis was used to measure Jak2 expression levels.

PAGE 79

79 CHAPTER 5 DISCUSSION Overview It has been 15 years since the Jak2 gene was cloned (1). Since then, a wealth of evidence has accumulated linking Jak2 to physiological and pa thophysiological processes. Jak2 has been implicated in cardiovascular disorders like atherosclerosis and cardiac ischemia-reperfusion injury. It has also been linke d to diabetic complications and cancer. These findings justify the need to study Jak2 function. However, while much progress has been made in trying to understand the factors that control Jak2 function, there is still much work to be done. Six of the eight tyrosine phosphorylation site s within Jak2 have been linked to regulatory mechanisms, and many more may still remain to be identified. Furthermore, an interesting development in the field came with the discovery of the first site of Jak2 serine phosp horylation in 2006 (88). This discovery opened up the possibility of other, nontyrosine regulatory sites in Jak2. With every new discovery, it becomes in creasingly more apparent that much more investigation is needed This work is an investigation into th e Jak2 kinase domain and the structure-function relationship that governs its func tion. Specifically, we looked at two sites within the kinase domain: Y972 and S1120. Both of these residues we re shown to be important for Jak2 kinase function. However, altering S1120 had a much more severe effect on Jak2 function than altering Y972. Thus, in terms of their re levance to larger physiological pr ocesses, we would expect to see differential effects. In this chapter, we will discuss the impact of these residues on Jak2 function and their potential roles in Jak2-dependent ROS pathologies. Tyrosine 972 and Jak2 Function As discussed in chapter 1, tyrosine autophos phorylation is a centra l process in Jak2dependent signaling. In Chapter 3, we describe our investigati on into Y972, a novel site of Jak2

PAGE 80

80 autophosphorylation. Through ms/ms mass spectrometr y, we confirmed that Y972 is a site of Jak2 autophosphorylation. We demonstrated that Y972 phosphorylation is crucial for Jak2 autophosphorylation. We also showed that the lo ss of phosphorylation at this site reduced Jak2 kinase function. The loss of Y972 phosphorylati on hindered several elemen ts of Jak2-dependent signaling, such as angiotensin II dependent Jak2 phosphorylation, growth hormone-dependent Jak2 activation, and growth hormone-dependent STAT1 nuclear translocation. We hypothesized that, while Y972 phosphorylation significantly affected the phosphorylation status of Y1007/Y1008, suboptimal Jak2 activati on that are not Y1007/Y1008 depe ndent, were left intact. Y972 and ROS Pathology The most important findings of our inves tigation into Y972 are 1) Y972 phosphorylation affects Jak2 kinase function, 2) Y972 phosphor ylation affects Jak2-dependent signal transduction, and 3) Y972-dependent effects are not irreversible. Within the context of ROSdependent pathologies, these three findings may have important consequences for Jak2 function. Cardiac Ischemia Reperfusion Injury Ischemia is a double-edged sword with respec t to the heart. Acute ischemic injury produces large amounts of ROS, which lead to cardiomyocyte loss through both apoptosis and necrosis. In contrast, preconditioni ng the heart by exposing it to shorter bouts of ischemia than those seen in acute injuries actu ally has a protective effect. Inte restingly, Jak2 plays a role in both of these processes. In fact Jak2 activation mirrors the role that ROS plays in the ischemic heart as both are protective w ithin the context of preconditioning and detrimental in acute ischemic injury. In Chapter 1, we hypothesized that Jak2 responds differe ntially to the dose of ROS in a given circumstance. Thus, it mediates protective effects for th e smaller doses of ROS associated with preconditioning while promoting apoptosis in response to the larger doses released by an acute ischemic injury.

PAGE 81

81 Tyrosine 972 may be beneficial as a therapeutic target for acut e cardiac I/R injury. It was previously shown that Jak2 mediates RO S-dependent apoptosis in response to hypoxia/reoxygenation (77). The loss of ki nase function incurred by inhibiting Y972 phosphorylation may hinder this apoptotic e ffect by delaying Jak2-dependent apoptotic signaling. Furthermore, Y972 phosphorylation wa s shown to be critical for maximal Jak2 activation. Given our hypothesis that Jak2 differentially responds to small and large ROS doses, the inhibition of Y972 phosphoryl ation may prevent Jak2 from b ecoming maximally activated in response to the larger ROS doses associated with acute I/R injury. Thus, the detrimental, Jak2dependent effects of acute I/ R injury may be attenuated by inhibiting Y972 phosphorylation. This inhibition incurs a loss of kinase func tion and an inability to achieve maximal Jak2 activation, both of which may hinder the progression of I/R injury. In contrast, inhibition of Y972 phosphorylation may hinde r cardiac preconditioning. The preconditioning phenomenon is based on pre-emptive gene expression. Cardiomyocytes respond to preconditioning stimuli by increas ing the expression of pro-survival genes. As mentioned in Chapter 1, Jak2 may mediate preconditioning th rough STAT3-dependent gene expression as STAT3 has been shown to upregulate cardi oprotective genes like metallothionein and manganese superoxide dismutase (75,76). The Y972-dependent loss of kinase function may hinder the ability of Jak2 to fully activate STAT3, thereby reducing cardioprotective gene expression. Not only has Jak2 been implicated in ischemic precondi tioning, but it has also been implicated in several ligand-dependent preconditioning mechanisms, like TNFand angiotensin II-dependent preconditioning (68,70). As the loss of Y972 phosphorylation has been shown to hinder several elements of Jak2-dependent signaling, we assert that the loss of Y972 phosphorylation would negatively affect li gand dependent forms of preconditioning.

PAGE 82

82 Diabetic Nephropathy Evidence suggests that Jak2 plays a significant role in several major events associated with the progression of diabetic nephropathy, incl uding increases in th e expression of TGF, fibronectin, and collagen ( 44-46). Furthermore, the AT1 receptor has been shown to mediate increases in TGFand fibronectin in diabetic nephropa thy. The effect of Y972 phosphorylation on angiotensin II-dependent Jak2 phosphorylation and Jak2 kinase function in general suggest that therapeutically targeting Y972 in diabetic nephropathy may be beneficial. By blocking Y972 phosphorylation, it may be possible to impede the angiotensin II-dependent increases in TGFand fibronectin, thus slowing the progression of diabetic nephropathy. The idea of inhibiting Y972 phosphorylation is a more attractive therapeutic strategy within the context of diabetic nephropathy than in that of cardiac ischemia-reperfusion injury. Diabetic nephropathy is a progr essive disease, as opposed to an acute event like cardiac I/R injury. The goal in treating this disease should be to stop aberra nt increases in Jak2-dependent signaling while leaving a signifi cant level of function intact in order to carry out the housekeeping functions of the cell. In Chapter 3, we demonstrated that while the loss of Y972 phosphorylation certainly affected Jak2 function, this loss did not eradicate enzymatic function. We proposed that Y972 primarily affected the maximally activated Jak2 conformation and that suboptimal activation states may be left intact These suboptimal activation states may be sufficient to carry out the housekeeping functions of Jak2-dependent signaling while inadequate to produce the pathological eff ects associated with diabetic nephropathy. Thus, inhibiting Y972 phosphorylation may slow the progression of di abetic nephropathy, al lowing more time for additional treatment.

PAGE 83

83 Atherosclerosis Atherosclerosis is a disease characterized by the development of vascular lesions. It shares attributes in common with both cardiac I/R injury and diabetic nephropathy. Like diabetic nephropathy, atherosclerosis is a progressive disease as vascular lesions develop over a lifetime. It is also similar to cardiac I/R injury in th at ROS can have differential signaling effects, depending on the dose encountered. Small to m oderate ROS doses are used in proliferative signaling pathways in vascular smooth muscle cells (55). However, as our lab has shown, larger doses can activate apoptotic pathwa ys (60). Interestingly, Jak2 ha s been shown to mediate both categories of responses (55,60). The idea of Y972 as a therapeutic target in atherosclerosis is attractive because of its progressive nature and the differential effects that ROS produce within the context of this disease. Tyrosine 972 phosphoryl ation appears to affect maxima l Jak2 activation while leaving suboptimal activation states inta ct. Thus, we propose a model in which Y972 inhibition prevents the apoptotic effects associated with larg e ROS doses while preserving suboptimal Jak2 activation states in order to carry out cellula r housekeeping functions. This approach would allow long-term treatment of this disease and could be administered at the first signs of atherosclerotic lesion development. The Y972-de pendent loss of kinase function may hinder the progression of atherosclerosi s by delaying Jak2-dependent signaling events. Tyrosine 972 inhibition may also be beneficial in advanced lesions, where it is necessary to maintain a certain level of signaling activity in or der to synthesize proteins necessa ry for plaque stability while avoiding vascular smooth muscle cell apoptosis which can destabilize fibrous plaques. S1120 and Jak2 Function Serine 1120 resides in the murine Jak2 kina se domain. Through computer modeling and structure-function investigations, we identified S1120 as an amino acid that is critical for Jak2

PAGE 84

84 function. In silico analysis suggests that S1120 may be a key stabilizing component in a fourhelix structural motif within the Jak2 kinase dom ain. Furthermore, an evaluation of the solvent accessibility of S1120 revealed that it is solvent accessible. Therefor e, it may be a potential site of serine phosphorylation within Jak2. For these reasons, we d ecided to investigat e the effect of mutating S1120 to alanine on Jak2 function. This is a relatively conserva tive mutation as it only removes the OH functional group from serine. Thus, any phosphorylative capacity and any structurally important hydrogen bonding interactions should be nullified by this mutation. We investigated S1120 from an intrinsic angl e and within the contex t of growth hormone signaling. Intrinsically, we found that S1120 is critical for Jak2 autophosphorylation. Furthermore, we determined that the S1120A point mutation severely hindered ligandindependent gene transcription. In a signaling context, S1120 was shown to be important for several aspects of growth hormone signali ng, including growth hormone-dependent Jak2 activation, SH2Bdependent activation and growth hormo ne-dependent gene transcription. The effects of S1120 on growth hormone depende nt signaling were the most noteworthy and pathologically relevant findings of this study. We will discuss their implications for growth hormone dependent disorders. S1120 and Growth Hormone-Dependent Pathophysiology Growth hormone signaling plays an important role in human health. A delicate balance must be achieved as pathological effects are a ssociated with both excessive and inadequate growth hormone signaling. For instance, gr owth hormone is essential during human development and a deficiency results in short st ature (87). In contra st, excessive amounts of growth hormone can lead to gigantism and several forms of cancer (87). Wh ile there is evidence to suggest that Jak2 plays a role in regulating the growth hormone /IGF-1 system, there is still work to be done in fully defining the role of Ja k2 in this signaling axis. For instance, while it is

PAGE 85

85 known that growth hormone acts through STAT5 to inhibit IGF1 binding protein (IGF1BP) expression, there is no direct evidence linki ng Jak2 to growth hormone-dependent STAT5 activation in this pathwa y (89). Furthermore, there is not a direct link between growth hormone dependent Jak2 activation and increases in IGF-1 e xpression. These questions must be addressed before a thorough understanding of the role of Jak2 in growth hormone dependent signaling and pathophysiology is obtained. A pharmacological inhibitor that is targeted toward S1120 may be a useful tool in investigating the role of Ja k2 in growth hormone-dependent signaling and pathophysiology. Similar to Y972, S1120 was found to be highly solvent-accessible, making it an attractive pharmacological target. Whether S1120 binds a phosphate group or forms an important hydrogen bond in the Jak2 kinase domain, a pharmacological inhibitor could mimic the effects of the S1120 mutation by blocking these in teractions, thus achieving th e S1120-dependent effects seen in Chapter 4. Serine 1120-dependent Jak2 inhibition may fac ilitate an understanding of the potential role of Jak2 in IGF-1 protein expression. In chapte r 4, we demonstrated that S1120 is critical for Jak2-dependent gene expression in response to growth hormone. Treating growth hormoneresponsive cells with an S1120 i nhibitor prior to challenging them with growth hormone would abolish the Jak2-dependent component of growth hormone-mediated gene expression. The IGF1 mRNA levels from these cells could be compar ed to cells that were not pretreated with inhibitor. Any differences found could be attri buted to Jak2-dependent gene transcription. The role of Jak2 in growth hormone dependent inhibition of IGF1BP expression could also be elucidated by S1120-dependent inhibition. Experiments evaluati ng this question would be set up in a manner similar to that described above. However, in this case, STAT5 activation and

PAGE 86

86 IGF1BP mRNA expression should be looked at. Since IGF1BP se questers IGF-1, a role for Jak2 in this proteins expression would be an im portant discovery as it would link Jak2 to the bioavailability of IGF-1. Y972 vs. S1120 In this dissertation, we describe the investig ations of two important amino acid residues in the Jak2 kinase domain. Each of these residues has its own unique role in the Jak2 kinase domain while also affecting some of the same pro cesses as well. Both residues reside in the Jak2 kinase domain (Fig. 1). A comparison of the tw o sites and their relevance to Jak2 function can provide valuable insight into Jak2 and the structure-function relations hip that is so crucial to its cellular role. (Table 1). In a comparison of Y972 and S1120, differences far outnumber similarities. The main similarity between Y972 and S1120 is that they both significantly affect Jak2 autophosphorylation. Beyond this in itial finding, the two residues are significantly different from one another. Whereas the Y972-dependent effects on Jak2 function app ear to be reversible, those dependent on S1120 are immutable and mo re severe. For example, while SH2Bwas able to restore the Y972-dependent loss of Y1007/Y1008 phosphorylation, it was not able to restore Y1007/Y1008 phosphorylation in the S1120A mutant. Furthermore, while the Y972F mutant displayed a loss of kinase function which was expected to affect its ability to drive ligand-independent gene expression, this deficit was clearly able to be overcome as the Y972F mutant achieved wild-type levels of ligand-i ndependent gene expressi on. In contrast, the S1120A mutant was not able to achieve wild-type levels of ligand-independent gene expression, indicating that the effect of this mutation was mu ch more severe on the in trinsic elements of Jak2 function. In chapter 3, we propos ed that the loss of Y972 phosphor ylation conferre d a stability deficit to the active Jak2 conf ormation. On the other hand, mutating S1120 destabilized the

PAGE 87

87 entire Jak2 protein as significantly more Jak2 S1120A plasmid was needed to produce a level of expression that was comparable to Jak2 wild-type protein. The differences between Y972-dependent a nd S1120-dependent effects on Jak2 function have significant implications for the manner in which Ja k2 autophosphorylation data is interpreted. As mentioned above, both of these residues were shown to be important for Jak2 total phosphorylation and Y1007/Y 1008 phosphorylation. The fact that there is such a disparity in the functional outcomes of mutating these re sidues suggests that the autophosphorylation level of Jak2 may not be a fully accurate indicator of the functional capacity of Jak2. As discussed in chapter 1, phosphotyrosines play many different roles in Jak2 function, including acting as both positive and negative functional regulators. The fact that the Jak2 Y972F mutant retained a level of basal function while displaying an almost to tal loss in tyrosine phos phorylation may reflect the balance between the positive and negative regula tors that were affected. In contrast, the S1120A mutation clearly affected the overall stab ility of the Jak2 protein and thus its loss of function may be more attributable to effects on structural stability than loss of tyrosine phosphorylation. Reflections The investigations into Y972 a nd S1120 are part of an ongoing effort to identify structural elements within Jak2 that are critical for its f unction. Both studies cont ribute to the body of knowledge pertaining the relationship between Ja k2 structure and its function. However, no study is perfect and hindsight is often the best tool for identifying missing elements. Thus, in retrospect we have identified a few elements that would be excelle nt additions to these investigations.

PAGE 88

88 Y972 The investigation into Y972 and its effect on Jak2 function unc overed a differential role for Y972 with respect to angiotensi n II and growth hormone depende nt Jak2 activation. In our conclusions, we asserted that this might be due to the inherent differences between cytokine and GPCR signaling. In retrospect, we should have developed this aspe ct of the investigation to a much greater extent. Specifically, we should ha ve looked at the role of Y972 phosphorylation in multiple cytokine and GPCR signaling pathways to determine if the effects we saw were specific to angiotensin II and growth hormone signaling or if they are truly representative of a differential role for Y972 phosphorylation with respect to GPCR and cytokine signaling. Furthermore, the Y972 investigation lacks pa rallelism with respect to the comparison of angiotensin II and growth hormone signaling. T hus, experiments looking at the role of Y972 phosphorylation in angiotensi n II dependent Y1007/Y1008 phosphor ylation and angiotensin II dependent STAT1 nuclear translocation would be va luable additions to this study as they would allow for more complete comparisons between thes e two signaling pathways Finally, this study would benefit from an evaluation of the role of Y972 phosphorylation in Jak2 dependent gene expression in these two pathways. This aim c ould be achieved by utiliz ing the fibroblast cell lines stably transfected with th e Jak2 Y972F construct. These el ements would greatly facilitate the characterization of the role of Y972 phosphorylation in Jak2 function. S1120 The investigation into S1120 lacks a mechanistic explanation for the S1120 dependent effects on Jak2 function. In our investigation, we hypot hesized that the S1 120-dependent effects that we saw may be due the effect of S1120 on Jak2 dimerization. The in silico data, which suggests that S1120 resides near the Jak2 dimer in terface partially supports this hypothesis. The

PAGE 89

89 S1120 investigation would greatly benefit from an evaluation of the role of S1120 in Jak2 dimerization, similar to the dimerization expe riment conducted in the Y972 investigation. Like the Y972 study, the S1120 study would grea tly benefit from a more comprehensive examination of the role of S1120 in Jak2 dependent signaling. For instan ce, the effect of S1120 on multiple Jak2-dependent signaling pathways w ould provide a more complete picture of the role of S1120 in Jak2 function. In conclusion, these two studies highlight the importance and complexity of the Jak2 structure/function relationship. Jak2 amino acids can have prof ound effects on Jak2 function. It is clear that more investig ation is needed in order to fully understand the role of autophosphorylation in Jak2 function. Only eight of the 49 Jak2 tyrosines have been identified as phosphotyrosines. However, as we discussed ab ove, our investigation in to Y972 suggests that there may be many more phosphotyrosine regulatory sites in Jak2 and it is the summation of their individual effects that governs Jak2 functio n. Additionally, non-tyrosine residues must be investigated as well. These residues may partic ipate in important interactions that maintain proper Jak2 function. The more we understand a bout the Jak2 structur e/function relationship, the more insight we will gain into its pivotal role in human health.

PAGE 90

90 Table 5-1. A comparison of S1120A-dependent and Y972F-dep endent effects on several categories of Jak2 function. The S1120A a nd Y972F mutations have similar impacts on Jak2 autophosphorylation. However, there is a significant divergence with respect to their effects on all other categories ex amined. The S1120A mutation is a much more severe structural ch ange as it has much stronger effects on Jak2 function.

PAGE 91

91 Figure 5-1. The relative locat ions of Y972 and S1120 in murine Jak2. Tyrosine 972 is highlighted in red and S1120 is highlighted in blue.

PAGE 92

92 LIST OF REFERENCES 1. Wilks, A.F., Harpur, A.G., Kurban, R.R., Ralph, S.J., Zurcher, G., and Ziemiecki, A. (1991) Mol. Cell. Biol 11 2057-2065 2. Ihle, J.N., Witthuhn, B., Tang, B., Yi., T., and Quelle, F.W. (1994) Baillieres. Clin. Haematol 7 17-48 3. Witthuhn, B.A., Quelle, F.W., Silvennoinen, O ., Yi, T., Tang, B., Miura, O., and Ihle, J.N. (1993) Cell 74 227-236 4. Argetsinger, L.S., Campbell, G.S., Yang, X., Witthuhn, B.A., Silvennoinen, O., Ihle, J.N., and Carter-Su, C. (1993) Cell 74 237-244 5. Silvennoinen, O., Witthuhn, B.A., Quelle, F.W., Cleveland, J.L., Yi, T., and Ihle, J.N. (1993) Proc. Natl. Acad. Sci. U.S.A 90 8429-8433 6. Watling, D., Guschin, D., Muller, M., Silv ennoinen, O., Witthuhn, B.A., Quelle, F.W., Rogers, N.C., Schindler, C., Star k, G.R., and Ihle, J.N. (1993) Nature 366 166-170 7. Rui, H., Kirken, R.A., and Farrar, W.L. (1994) J. Biol. Chem 269 5364-5368 8. Narazaki, M., Witthuhn, B.A., Yoshida, K., S ilvennoinen, O., Yasukawa, K., Ihle, J.N., Kishimoto, T., and Taga, T. (1994) Proc. Natl. Acad. Sci. U.S.A. 91 2285-2289 9. Rodriguez-Linares, B., and Watson, S.P. (1994) FEBS Lett 352 335-338 10. Marrero, M.B., Schieffer, B., Paxton, W.G., Heerdt, L., Berk, B.C., Delafontaine, P., and Bernstein, K.E. (1995) Nature 375 247-250 11. Simon, A.R., Rai, U., Fanburg, B. L., and Cochran, B.H. (1998) Am. J. Physiol 275 C1640-1652 12. Kralovics, R., Passamonti, F., Buser, A.S., Te o, S.S., Tiedt, R., Passweg, J.R., Tichelli, A., Cazzola, M., and Skoda, R.C. (2005) N. Engl. J. Med 352 1779-1790 13. Saharinen, P., Takaluoma, K ., and Silvennoinen, O. (2000) Mol. Cell. Biol 20 33873395 14. Sandberg, E.M., VonDerLinden, D., Ostr ov, D.A., and Sayeski, P.P. (2004) Mol. Cell. Biochem 265 161-169 15. Frenzel, K., Wallace, T.A., McDoom, I., Xia o, H.D., Capecchi, M.R., Bernstein, K.E., and Sayeski, P.P. (2006) Exp. Cell. Res 312 2735-2744 16. VonDerLinden, D., Ma X., Sandberg, E.M., Gern ert, K., Bernstein, K.E., and Sayeski, P.P. (2002) Mol. Cell. Biochem 241 87-94

PAGE 93

93 17. Levine, R.L., and Wernig, G. (2006) Hematology Am. Soc. Hematol. Educ. Program 510 233-239 18. Kralovics, R., Teo, S.S., Buser, A.S., Brutsche, M., Tiedt, R., Tichelli, A., Passamonti, F., Pietra, D., Cazzola, M., and Skoda, R.C. (2005) Blood 106 3374-3376 19. Feng, J., Witthuhn, B.A., Matsuda, T., Kohlhuber, F., Kerr, I.M., and Ihle, J.N. (1997) Mol. Cell. Biol. 17 2497-2501 20. Ungureanu, D., Saharinen, P., Junttila, I., Hilton, D.J., and Silv ennoinen, O. (2002) Mol. Cell. Biol. 22 3316-3326 21. Kurzer, J.H., Argetsinger, L.S., Zhou, Y.J., K ouadio, J.L., OShea, J.J., and Carter-Su, C. (2004) Mol. Cell. Biol 24 4557-4570 22. Argetsinger, L.S., Kouadio, J.L., Steen, H., Stensballe, A., Jensen, O.N., and Carter-Su, C. (2004) Mol. Cell. Biol 24 4955-4967 23. Godeny, M.D., Sayyah, J., VonDerLinden, D., Johns, M., Ostrov, D.A., Caldwell-Busby, J., and Sayeski, P.P. (2007) Cell. Signal 19 600-609 24. Funakoshi-Tago M, Pelletier, S., Matsuda, T., Parganas, E, and Ihle, J.N. (2006) EMBO J. 25, 4763-4772 25. Ihle, J.N., Witthuhn, B.A., Quelle, F.W., Yamamo to, K., Thierfelder, W.E., Kreider, B., and Silvennoinen, O. (1994) Trends Biochem. Sci 19 222-227 26. Ihle, J.N. (1994) Proc. Soc. Exp. Biol. Med 206 268-272 27. Brooks, A.J., Wooh, J.W., Tunny, K. A., Waters, M.J., (2007) Int. J. Biochem. Cell. Biol. Epub ahead of print PMID: 17888716 28. Fong, T.M. (1996) Cell. Signal 8 217-224 29. Ferrand, A., Kowalski-Chauvel, A., Bertrand, C. Escrieut, C., Mathieu, A., Portolan, G., Pradayrol, L., Fourmy, D., Dufresne, M., and Seva, C. (2005) J. Biol. Chem 280 1071010715 30. Luchtefeld, M., Drexler, H ., and Schieffer, B. (2001) Biochem. Biophys. Res. Commun 280 756-760. 31. Frank, G.D., Saito, S., Motley, E.D., Sasaki, T., Ohba, M., Kuroki, T., Inagami, T., Eguchi, S. (2002) Mol. Endocrinol 16 367-377 32. Ali, S., Nouhi, Z., Chughtai, N., Ali, S. (2003) J. Biol. Chem 278 52021-52031 33. Chen, J., Yu, W.M., Bunting, K.D., and Qu, C.K. (2004) Oncogene 23 3659-3669 34. Maures, T.J., Kurzer, J.H., and Carter-Su, C. (2007) Trends Endocrinol. Metab 18, 38-45

PAGE 94

94 35. Kurzer, J.H., Saharinen, P., Silve nnionen, O., and Carter-Su, C. (2006) Mol. Cell. Biol 26 6381-6394 36. Starr, R., Willson, T.A., Viney, E.M., Murray, L.J., Rayner, J.R., Jenkins, B.J., Gonda, T.J., Alexander, W.S., Metcalf, D ., Nicola, N.A., Hilton, D.J. (1997) Nature 387 917921 37. Torsoni, M.A., Carvalheira, J.B., Calegari V.C., Bezerra, R.M., Saad, M.J., Gontijo, J.A., and Velloso, L.A. (2004) J. Endocrinol 181 117-128 38. Martens, N., Uzan, G., Wery, M., Hooghe, R ., Hooghe-Peters, E.L. and Gertler, A. (2005) J. Biol. Chem. 280 13817-13823 39. Cross, J.V., and Templeton, D.J. (2006) Antioxid. Redox. Signal. 8 1819-1827 40. McDoom, I., and Sayeski, P.P. (2007) Current Chemical Biology 1 303-310 41. Huang, J.S., Guh, J.Y., Yang, M.L., Lai, Y.H., Chen, H.C., and Chuang, L.Y. (1999) Biochem. J. 342, 231-238 42. Wang, X., Shaw, S., Amiri, F., Eato n, D., and Marrero, M.B. (2002) Diabetes 51 35053509 43. Amiri, F., Shaw, S., Wang, X., Tang, J., Walle r, J., Eaton, D., and Marrero, M.B. (2002) Kidney Int 61 1605-1616 44. Tuttle, K., Johnson, E., Cooney, S., Anderber g, R.J., Johnson, E.K., Clifton, G.D., and Meek, R.L. (2005) Kidney Int 67 953-968 45. Dan, Q., Wong, R.L., Yin, S., Chung, S.K ., Chung, S.S., and Lam, K.S. (2004) Nephron Exp Nephrol 98 e89-99 46. Fukami, K., Ueda, S., Yamagishi, S., Kato, S ., Inagaki, Y., Takeuchi, M., Motomiya, Y., Bucala, R., Iida, S., Tamaki, K., Imaizu mi, T., Cooper, M., and Okuda, S. (2004) Kidney Int 66, 2137-2147 47. Lee, C., Guh, J.Y., Chen, H.C., Hung, W.C., Yang, Y.L., and Chuang, L.Y. (2005) J Cell Biochem 95, 281-292 48. Cai, L., Li, W., Wang, G., Guo, L ., Jiang, Y., and Kang, Y.J. (2002) Diabetes 51 19381948 49. Rota, M., Le Capitaine, N., Hosoda, T., Boni A., De Angelis, A., Padin-Iruegas, M.E., Esposito, G., Vitale, S., Urbanek, K., Casarsa, C., Giorgio, M., Luscher, T.F., Pelicci, P.G., Anversa, P., Leri, A ., and Kajstura, J. (2006) Circ. Res 99 42-52 50. Aragno, M., Mastrocola, R., Medana, C., Cata lano, M.G., Vercellinatto, I., Danni, O., and Boccuzzi, G. (2006) Endocrinology 147 5967-5974

PAGE 95

95 51. Modesti, A., Bertolozzi, I., Gamberi, T ., Marchetta, M., Lumachi, C., Coppo, M., Moroni, F., Toscano, T., Lucchese, G., Ge nsini, G.F., and Modesti, P.A. (2005) Diabetes 54 394-401 52. Privratsky, J.R., Wold, L.E., Sowers, J.R., Quinn, M.T., and Ren, J. (2003) Hypertension 42, 206-212 53. Lilly, L.S. (1998) Pathophysiology of Heart Disease 2nd Ed, Williams & Wilkins, Baltimore, MD 54. Sung, H., Eskin, S., Sakurai, Y., Yee, A., Kataoka, N., and McIntire, L. (2005) Ann Biomed Eng 33 1546-1554 55. Madamanchi, N., Li, S., Patterson, C., and Runge, M. (2001) Arterioscler. Thromb. Vasc. Biol 21 321-326 56. Madamanchi, N., Li, S., Patterson, C., and Runge, M. (2001) J. Biol. Chem 22 1891518924 57. Li, P.G., Xu, J.W., Ikeda, K., Kobayakawa, A., Kayano, Y., Mitani, T., Ikami, T., and Yamori, Y. (2005) Hyperten. Res 28 369-377 58. Hsieh, C., Yen, M., Yen, C., and Lau, Y. (2001) Cardiovasc. Res 49 135-145 59. Maziere, C., Conte, M., and Maziere, J. (2001) Free Radic Biol Med 31 1334-1340 60. Sandberg, E., and Sayeski, P. (2004) J. Biol. Chem 279 34547-34552 61. Tawfik, A., Jin, L., Banes-Berc eli, A., Caldwell, R.B., Ogbi S., Shirley, A., Barber, D., Catravas, J.D., Stern, D.M., Fulton, D., Ca ldwell, R.W., and Marrero, M.B. (2005) Vascul. Pharmacol 43 320-326 62. Vianna, H.R., Cortes, S.F., Ferreira, A.J., Capettini, L.S., Schmitt, M., Almeida, A.P., Massensini, A.R., and Lemos, V.S. (2006) J. Planta. Med 72 300-303 63. Bansal, P., Gupta, S.K., Ojha, S.K., Nandave, M., Mittal, R., Kumari, S., and Arya, D.S. (2006) Mol. Cell. Biochem 289 1-9 64. Shen, M., Jia, G., Wang, Y, and Ma, H. (2006) Vascul. Pharmacol 45 122-126 65. Guaiquil, V., Golde, D., Beckles, D., Mascareno, E., and Siddiqui, M. (2004) Free Radic. Biol. Med. 37 1419-1429 66. Xuan, Y.T., Guo, Y., Han, H., Zhu, Y., and Bolli, R. (2001) Proc Natl Acad Sci 98 90509055 67. Das, D.K., Maulik, N., Sato, M., and Ray, P.S. (1999) Mol. Cell. Biochem 196 59-67

PAGE 96

96 68. Lecour, S., Suleman, N., Deuchar, G.A., So mers, S., Lacerda, L., Huisamen, B., and Opie, L.H. (2005) Circulation 112 3911-3918 69. Lecour, S., Rochette, L., and Opie, L. (2005) Cardiovasc. Res 65 239-243 70. Das, S., Engelman, R.M., Maulik, N., and Das, D. (2006) Cell. Biochem. Biophys 44 103-110 71. Schieffer, B., Luchtefeld, M., Braun, S., Hilf iker, A., Hilfiker-Klein er, D., and Drexler, H. (2000) Circ. Res 87 1195-1201 72. Dawn, B., Xuan, Y.T., Guo, Y., Rezazadeh, A., Stein, A.B., Hunt, G., Wu, W.J., Tan, W., and Bolli, R. (2004) Cardiovasc. Res 64 61-71 73. Gross, E.R., Hsu, A.K., and Gross, G.J. (2006) Am. J. Physiol. Heart Circ. Physiol 291 H827-H834 74. Ueda, K., Takano, H., Hasegawa, H., Niitsuma, Y., Qin, Y., Ohtsuka, M., and Komuro, I. (2006) Arterioscler. Thromb. Vasc. Biol 26 e108-113 75. Oshima, Y., Fujio, Y., Nakanishi, T., Itoh, N., Yamamoto, Y., Negoro, S., Tanaka, K., Kishimoto, T., Kawase, I., and Azuma, J. (2005) Cardivasc. Res 65 428-435 76. Negoro, S., Kunisada, K., Fujio, Y., Funamoto, M., Darville, M., Eizirik, D., Osugi, T., Izumi, M., Oshima, Y., Nakaoka, Y., Hirota, H., Kishimoto, T., and Yamauchi-Takihara, K. (2001) Circulation 104 979-981 77. Mascareno, E., Beckles, D.L ., and Siddiqui, M.A.Q. (2005) Vascul. Pharmacol 43 327335 78. Tritto, I., DAndrea, D., Eramo, N., Scognami glio, A., De Simone, C., Violante, A., Esposito, A., Chiariello, M ., and Ambrosio, G. (1997) Circ. Res 80 743-748 79. Lecour, S., Owira, P., and Opie, L. (2006) Life Sci 78 1702-1706 80. Kabsch, W., and Sander, C. (1983) Biopolymers 22 2577-2637 81. Ma, X., and Sayeski, P.P. (2004) Protein Expr. Purif 35 181-189 82. Sandberg, E.M., Wallace, T.A., Godeny, M.D ., Vonderlinden, D., and Sayeski, P.P. (2004) Cell. Biochem. Biophys. 41 207-232 83. Ali, M.S., Sayeski, P.P., Safavi, A., Lyles, M., and Bernstein, K.E. (1998) Biochem. Biophys. Res. Commun. 249, 672-677 84. Ali M.S., Sayeski, P.P., and Bernstein, K.E. (2000) J. Biol. Chem 275 15586-15593 85. Chatti, K., Farrar, W.L., and Duhe, R.J. (2004) Biochemistry 14 4272-4283

PAGE 97

97 86. Sayeski, P.P., Ali, M.S., Frank, S.J., and Bernstein, K.E. (2001) J. Biol. Chem. 276 10556-10563 87. Ayuk, J., and Sheppard, M. (2006) Postgrad. Med. J. 82 24-30 88. Mazurkiewicz-Munoz, A.M., Argetsinger, L.S ., Kouadio, J.L., Stensballe, A., Jensen, O.N., Cline, J.M., and Carter-Su, C. (2006) Mol. Cell. Biol 26 4052-4062 89. Ono, M., Chia, D.J., Merino-Martinez, R., Flores-Morales, A., Unterman, T.G., and Rotwein, P. (2007) Mol. Endocrinol 21 1443-1457

PAGE 98

98 BIOGRAPHICAL SKETCH Issam McDoom was born in Jeddah, Saudi Arab ia. He and his family moved to Miami, Florida in 1986. He lived in Miami for eight ye ars and then moved to Ja cksonville, Florida in 1994. Issam came to the University of Florida as an undergraduate in 1997. He studied History and Zoology. It was at the University of Flor ida College of Medicine that Issam acquired his interest in research by working as a laboratory te chnician in several laboratories. In 2003, Issam entered the University of Florida Interdisciplinar y Program in Biomedical Sciences. He trained in the laboratory of Dr. Peter P. Sayeski in the department of P hysiology and Functional Genomics.