<%BANNER%>

The PAQR Family of Membrane Proteins in Saccharyomyces cerevisiae

Permanent Link: http://ufdc.ufl.edu/UFE0021977/00001

Material Information

Title: The PAQR Family of Membrane Proteins in Saccharyomyces cerevisiae Their Role in the Signal Transduction Pathway of FET3 and Flow Cytometric Analysis of AdipoR1 Binding to Adiponectin and TNF-Alpha
Physical Description: 1 online resource (184 p.)
Language: english
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

Subjects / Keywords: adiponectin, iron, izh, paqr, progesterone, zinc
Chemistry -- Dissertations, Academic -- UF
Genre: Chemistry thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: The yeast Saccharomyces cerevisiae contains a family of zinc responsive membrane proteins named IZH1-4 (Implicated in Zinc Homeostasis), which belong to a superfamily of proteins known as the PAQRs (Progestin AdipoQ Receptor). These putative membrane proteins contain at least seven transmembrane domains. The PAQR family of proteins consists of known osmotin, adiponectin and progesterone receptors that are distantly related to alkaline ceramidases. The objective of this study is to characterize the role of the PAQR family in lipid mediating signaling. We have evidence that the PAQR family, when overexpressed or in the presence of their specific ligand, represses ZRT1, a high affinity zinc transporter gene, and FET3, a high affinity iron transporter gene, suggesting a role for these proteins in zinc- and iron-uptake. The repression of these genes requires a sphingoid-base dependent signal transduction pathway involving Ras-cAMP/PKA and the Snf1p AMP-dependent kinase. This pathway ultimately governs a competition between the Nrg1p/Nrg2p and Msn2p/Msn4p transcription factors for binding sites in the promoter region of ZRT1 and FET3. We also found that human AdipoR1 represses FET3 with the addition of adiponectin. Furthermore, the addition of TNF-? partially alleviates repression by AdipoR1. However, when adiponectin and TNF-? were added together, FET3 repression was not alleviated. Finally, we examined labeled-adiponectin and TNF-? binding to cells overexpresing AdipoR1. Bound TNF-? levels fell when unlabeled adiponectin was added; however, the addition of unlabeled TNF-? displaced labeled-adiponectin. Thus, adiponectin and TNF-? compete for binding to yeast cells expressing AdipoR1.
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: Lyons, Thomas J.
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: UFE0021977:00001

Permanent Link: http://ufdc.ufl.edu/UFE0021977/00001

Material Information

Title: The PAQR Family of Membrane Proteins in Saccharyomyces cerevisiae Their Role in the Signal Transduction Pathway of FET3 and Flow Cytometric Analysis of AdipoR1 Binding to Adiponectin and TNF-Alpha
Physical Description: 1 online resource (184 p.)
Language: english
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

Subjects / Keywords: adiponectin, iron, izh, paqr, progesterone, zinc
Chemistry -- Dissertations, Academic -- UF
Genre: Chemistry thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: The yeast Saccharomyces cerevisiae contains a family of zinc responsive membrane proteins named IZH1-4 (Implicated in Zinc Homeostasis), which belong to a superfamily of proteins known as the PAQRs (Progestin AdipoQ Receptor). These putative membrane proteins contain at least seven transmembrane domains. The PAQR family of proteins consists of known osmotin, adiponectin and progesterone receptors that are distantly related to alkaline ceramidases. The objective of this study is to characterize the role of the PAQR family in lipid mediating signaling. We have evidence that the PAQR family, when overexpressed or in the presence of their specific ligand, represses ZRT1, a high affinity zinc transporter gene, and FET3, a high affinity iron transporter gene, suggesting a role for these proteins in zinc- and iron-uptake. The repression of these genes requires a sphingoid-base dependent signal transduction pathway involving Ras-cAMP/PKA and the Snf1p AMP-dependent kinase. This pathway ultimately governs a competition between the Nrg1p/Nrg2p and Msn2p/Msn4p transcription factors for binding sites in the promoter region of ZRT1 and FET3. We also found that human AdipoR1 represses FET3 with the addition of adiponectin. Furthermore, the addition of TNF-? partially alleviates repression by AdipoR1. However, when adiponectin and TNF-? were added together, FET3 repression was not alleviated. Finally, we examined labeled-adiponectin and TNF-? binding to cells overexpresing AdipoR1. Bound TNF-? levels fell when unlabeled adiponectin was added; however, the addition of unlabeled TNF-? displaced labeled-adiponectin. Thus, adiponectin and TNF-? compete for binding to yeast cells expressing AdipoR1.
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: Lyons, Thomas J.
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: UFE0021977:00001


This item has the following downloads:


Full Text

PAGE 1

1 THE PAQR FAMILY OF MEMBRANE PROTEINS IN Saccharomyces cerevisiae: THEIR ROLE IN SIGNAL TRANSDUCTION AND FLOW CYTOMETRIC ANALYSIS OF AdipoR1 BINDING TO ADIPONECTIN AND TNF-ALPHA By BRIAN RICHARD KUPCHAK 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 Brian Richard Kupchak

PAGE 3

3 To my family, whose faith and strength are my driving force to deal with adversity, and th e challenges in my life.

PAGE 4

4 ACKNOWLEDGMENTS It is m y honor and pleasure to thank my parents (Richard and Jane) and my sister (Melissa) for their support thr oughout the struggles of my studi es. I thank my mother for instilling in me the understanding that educati on was important, and my father for encouraging me to do everything at the best of my ability and never give up. I must express thanks to Dr. Tom Ly ons, my advisor, for his for openness, encouragement, support, and guidance over the years. His knowledge, integrity, infectious enthusiasm for research, work ethic, and sincere attitude toward life provide an extraordinary example for me both in science and life. I am pr ivileged to have been a recipient of his time and support. Next, I must gratefully acknowledge the effort of the fine educators on my dissertation committee: Dr. Jon Stewart, Dr. Nicole Horenstein, Dr. Brajter-Toth, and especially Dr. Robert Cousins for allowing me to use his fluorescence micr oscope. I have learned so much from these individuals throughout the years as teachers in the classroom and mentors in journal club and seminars. Without their guidance and support, I would not be the scientist I am today. I also need to thank the members of the Flow Cytometry Core (Neal Benson and Steve McClell) for their help and guidance the past few years. They were alwa ys generous in allowing me to utilize their facilities and guide me through my problems. I need to say thank you to the members of my group both present and past (Lisa, Nancy, Stephanie, Julie, Ibon, Jessica, Li dia, Marilee, Kim, and Matt) for their involvement in my studies and discussions, for their friendship, and their assistance over the past five years. Special thanks go to Dr. Willard Harrison and Jordan Math ias for their help in revising this dissertation. Finally, there is not enough time and space to th ank all of my friends. I am blessed to have come in contact with such kind and generous individuals in my life. I would especially thank Jeff Mills, Jason Sico, and Matthew Soulsby who I always looked up to them and tried to

PAGE 5

5 emulate their academic success. Without their pursuit of knowledge and their friendship I would have never be in the position I am in today.

PAGE 6

6 TABLE OF CONTENTS page ACKNOWLEDGMENTS...............................................................................................................4LIST OF FIGURES.........................................................................................................................9ABSTRACT...................................................................................................................................12CHAPTER 1 INTRODUCTION..................................................................................................................14Iron and Zinc Homeostasis..................................................................................................... 14Identification of the IZH s........................................................................................................15Structural Features............................................................................................................ ......16Class I, II, and III Receptors................................................................................................. ..18NRG and MSN.................................................................................................................... ...23AMPK.....................................................................................................................................25PKA........................................................................................................................................26PDK (PKH1 and PKH2).........................................................................................................26Sphingolipids..........................................................................................................................27Yeast as a Model Organism.................................................................................................... 30Summation..............................................................................................................................322 THE EFFECT OF IZH2 ON FET3 AND ZRT1 .....................................................................45Introduction................................................................................................................... ..........45Materials and Methods...........................................................................................................47Yeast Strains and Plasmids.............................................................................................. 47lacZ Reporter Constructs.................................................................................................48 Galactosidase and Ferroxidase Assays.......................................................................49Ferrozine Plates...............................................................................................................50Protein Isolation.............................................................................................................. .50Gel Electrophoresis......................................................................................................... 50Western Blot....................................................................................................................50Results.....................................................................................................................................51Effect of IZH2 Overexpression on the ZRT1 Promter..................................................... 51Effect of IZH2 Overexpression on the FET3 Promoter...................................................52Msn2p and Msn4p Activates FET3 .................................................................................53Nrg1p and Nrg2p Represses FET3 ..................................................................................54Discussion...............................................................................................................................55Repression of FET3 and ZRT1 Involves the CCCTC Regulatory Element.................... 55Nrg1p/Nrg2p and Msn2p/Msn4p Regulate FET3 ...........................................................58Nrg1p and Nrg2p Negatively Regulates FET3 ...............................................................59Msn2p/Msn4p and Nrg1p/Nrg2p are Epistatic................................................................ 59Aft1p Activation is not Aff ected by Mutant Strains....................................................... 60Summation..............................................................................................................................60

PAGE 7

7 3 THE SIGNALING OF FET3 REPRESSION BY IZH2 OVEREXPRESSION .................... 78Introduction................................................................................................................... ..........78Materials and Methods...........................................................................................................78Yeast Strains and Plasmids.............................................................................................. 78 -Galactosidase and Ferroxidase Assays........................................................................ 79Results.....................................................................................................................................79PKA and Ras-cAMP is involved in FET3 regulation.....................................................79AMPK is involved in FET3 Regulation..........................................................................80FET3 Regulation Requires PKH1 and PKH2 .................................................................80Mutant Strains do not Effect Aft1p Regulation...............................................................81Discussion...............................................................................................................................81AMPK..............................................................................................................................81PKAs Role in IZH2 -Dependent Repression................................................................... 82Pkh1p/Pkh2p as a Master Regulator................................................................................83Summation..............................................................................................................................834 PAQR-DEPENDENT REPRESSION OF FET3 IS MEDIATED BY SPHINGOLIPIDS .... 94Introduction................................................................................................................... ..........94Materials and Methods...........................................................................................................95Yeast Strains and Plasmids.............................................................................................. 95 -Galactosidase and Ferroxidase Assays........................................................................ 96Results.....................................................................................................................................96Exogenous and Endogenous Sphingoid Bases Repress FET3 ........................................96Exogenous and Endogenous Sphingoid Bases Mimic the Effect of IZH2 Overexpression............................................................................................................97IZH2 Overexpression is Affec ting Sphingoid Metabolism.............................................97Accumulation of Sphingoid Bases Effects FET3 ............................................................98Endogenous Alkaline Ceramidase are not Required for FET3 Signaling....................... 98Discussion...............................................................................................................................99Effect of Sphingoid Bases on FET3 ................................................................................99Izh2p-Dependent Signaling is Alleviated By Disruption of Sphingoid Base Synthesis....................................................................................................................100Accumulation of Sphingoid Bases Represses FET3 .....................................................100Inhibition of Ceramidase Activity Alleviate IZH s and YPC1 Effect on FET3 .............102FET3 Signaling does not Require Endogenous Alkaline Ceramidases......................... 103Summation............................................................................................................................1045 PAQRS REPRESS FET3 IN SIMIL AR MANNER AS IZH2 .............................................124Introduction................................................................................................................... ........124Materials and Methods.........................................................................................................126Yeast Strains and Plasmids............................................................................................ 126 Galactosidase and Ferroxidase Assays.....................................................................126Results...................................................................................................................................127

PAGE 8

8 Effect of Human PAQRs on FET3 Expression ............................................................. 127AdipoR2 and mPR are Activated by its Putative Ligands.......................................... 127PAQRs affect FET3 Expression.................................................................................... 128Discussion.............................................................................................................................129PAQRs Repress FET3 Expression................................................................................ 129Effect of Class II Homologues on FET3 Expression.................................................... 130Summation............................................................................................................................1326 THE EFFECTS OF ADIPONECTIN AND TUMOR NECROSIS FACTOR ON ADIPOR1 .............................................................................................................................146Introduction................................................................................................................... ........146Materials and Methods.........................................................................................................147Yeast Strains and Plasmids............................................................................................ 147 -Galactosidase Assays................................................................................................148Spheroplast Formation................................................................................................... 148Addition of Fluorinated Ligands................................................................................... 148Flow Cytometry............................................................................................................. 149Results...................................................................................................................................149Effect of AdipoR1 on FET3 Expression....................................................................... 150Adiponectin Binds to AdipoR1 on Yeast Spheroplasts................................................. 150TNFbinds to AdipoR1 on Spheroplasts.................................................................... 150Adiponectin, rather than TNF, prefers binding to AdipoR1......................................151Discussion.............................................................................................................................151Effect of AdipoR1 on Fet3 expression.......................................................................... 151Adiponectin binding to Ad ipoR1 on spheroplasts.........................................................152TNFbinding to AdipoR1 on spheroplasts................................................................. 153Adiponectin, rather than TNF, prefers binding to AdipoR1......................................154Summation............................................................................................................................155APPENDIX A YEAST STRAINS UTILIZED THOUGHOUT THIS STUDY ..........................................163B PRIMERS USED FOR CLONING......................................................................................165C MEDIA ABREVIATIONS AND COMPOSITION............................................................. 168LIST OF REFERENCES.............................................................................................................169BIOGRAPHICAL SKETCH.......................................................................................................184

PAGE 9

9 LIST OF FIGURES Figure page 1-1 The transcription factor Zap1p binds to the zinc response elem ent (ZRE) on the promoter region of genes in Saccharomyces cerevisiae during periods of low zinc......... 34 1-2 A Saccharomyces cerevisiae cell showing the zinc transporters. ...................................... 34 1-3 The transcription factor Aft1p binds to the iron response elem ent (FeRE) on the promoter region of genes in Saccharomyces cerevisiae during periods of low iron......... 35 1-4 A Saccharomyces cerevisiae cell showing the iron transporters. ...................................... 35 1-5 Multiple sequence alignment of PAQR proteins. Areas highlighted are highly conserved m otifs............................................................................................................... .36 1-6 Phylogenetic tree displaying IZH and hum an PAQR proteins. .......................................37 1-7 Topology predicted for all members of the PAQRs with the locations and the consensus sequences of the three highly conserved m otifs. s...........................................38 1-8 Kyte-doolittle plot of PAQR and PAQR-like proteins.. .................................................... 39 1-9 Adiponectin domains and forms........................................................................................ 40 1-10 Synthesis of progestins, glucocortic oids, m ineralocorticoids, androgens, and estrogens. All steroids synthesized are from pre gnenolone via cholesterol..................... 41 1-11 A model for the mechanism of FET3 by PAQR overexpression. .................................... 42 1-12 The AMPK complex and activators. ................................................................................ 43 1-13 Saccharomyces cerevisiae genes IZH 1 and IZH4 are divergently transcribed from PKH1 and PKH2 ................................................................................................................43 1-14 Sphingolipid biosynthesis in Saccha romyces cerevisiae ................................................44 2-1 A model for the mechanism of FET3 by IZH2 overexpression. ......................................61 2-2 Repression of ZRT1 by IZH2 -overexpression. ................................................................. 62 2-3 Repression of FET3 by IZH2 -overexpression. ................................................................ 65 2-4 Izh2p-3HA expression. .................................................................................................. 69 2-5 MSN2 and MSN4 positively regulate FET3 ....................................................................70 2-6 Nrg1p and Nrg2p are esse ntial for repression. ................................................................. 72

PAGE 10

10 2-7 Aft1p responds normally to iron-defeci ency in most m utant strains. .............................. 76 3-1 A model for the mechanism of FET3 by IZH2 overexpression. ......................................85 3-2 PKA and Ras-cAMP is involved in FET3 regulation. .....................................................86 3-3 AMPK is involved in FET3 regulation. .............................................................................88 3-4 FET3 regulation requires PKH1 and PKH2 ....................................................................91 3-5 Aft1p responds normally to iron-defeci ency in most m utant strains. .............................. 92 4-1 Phylogenetic analysis of PAQRs. ..................................................................................105 4-2 Topology predicted for all members of the PAQRs and alkaline ce ram idases with the locations and the consensus sequences of the three highly conserved motifs. .............. 106 4-3 A model for the mechanism of FET3 by IZH2 overexpression. ....................................107 4-4 Exogenous and endogenous sphingoid bases affect FET3 ...........................................108 4-5 Sphingoid bases cause an IZH -dependent repression of FET3 .......................................110 4-6 Inhibitors of sphingolipid synthesis and breakdown. ....................................................113 4-7 IZH overexpression is affecti ng sphingoid m etabolism...................................................115 4-8 Accumulation of sphingoid bases effects FET3 ...........................................................118 4-9. Inhibitor of ceramide breakdown. .................................................................................. 120 4-10 Involvement of alkaline ceramidas es in Izh2p-dependent signaling. ............................121 4-11 Role of sphingolipids in IZH -dependent signaling of FET3. .........................................123 5-1. Class I, II, and III PAQRs repress FET3 transcr iption : .................................................133 5-2 Nrg1p and Nrg2p effect on PAQR-dependent FET3 repression: ..................................136 5-3 AMPK in FET 3 repression: ...........................................................................................138 5-4 Sip3p in FET3 repression: .............................................................................................140 5-5 The role of Ras2p and PKA in PAQR-dependent repression: .......................................142 5-6 PDK is essential for human PA QR-dependent repression: .......................................... 144 6-1 AdipoR1 represses FET 3 transcription: .........................................................................156 6-2 Adiponectin binds to AdipoR1 on spheroplasts: ...........................................................157

PAGE 11

11 6-3 TNFbinds to AdipoR1 on spheroplasts: ....................................................................159 6-4 Competition of binding to Adi poR1 between adiponectin an d TNF: ........................161

PAGE 12

12 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 THE PAQR FAMILY OF MEMBRANE PROTEINS IN Saccharomyces cerevisiae: THEIR ROLE IN THE SIG NAL TRANSDUCTION PATHWAY OF FET3 AND FLOW CYTOMETRIC ANAL YSIS OF AdipoR1 BINDING TO ADIPONECTIN AND TNF-ALPHA By Brian Richard Kupchak May 2008 Chair: Thomas J. Lyons Major: Chemistry The yeast Saccharomyces cerevisiae contains a family of zinc responsive membrane proteins named IZH1-4 (Implicated in Zinc Homeostasis) which belong to a superfamily of proteins known as the PAQRs (Progestin AdipoQ Receptor). These putative membrane proteins contain at least seven transmembrane domains. The PAQR family of prot eins consists of known osmotin, adiponectin and progesterone receptors that are distantly related to alkaline ceramidases. The objective of this study is to char acterize the role of the PAQR family in lipid mediating signaling. We have evidence that th e PAQR family, when overexpressed or in the presence of their specific ligand, represses ZRT1 a high affinity zinc transporter gene, and FET3 a high affinity iron transporter gene suggesting a role for these prot eins in zincand iron-uptake. The repression of these genes requires a sphingo id-base dependent signal transduction pathway involving Ras-cAMP/PKA and the Snf1p AMP-depe ndent kinase. This pathway ultimately governs a competition between the Nrg1p/Nrg2p and Msn2p/Msn4p transcription factors for binding sites in the promoter region of ZRT1 and FET3 We also found that human AdipoR1 represses FET3 with the addition of adiponectin. Furthermore, the addition of TNFpartially alleviates repressi on by AdipoR1. However, when

PAGE 13

13 adiponectin and TNFwere added together, FET3 repression was not alleviated. Finally, we examined labeled-adiponectin and TNFbinding to cells overe xpresing AdipoR1. Bound TNFlevels fell when unlabeled adiponectin was added; however, the addi tion of unlabeled TNFdisplaced labeled-adiponectin. Thus, adiponectin and TNFcompete for binding to yeast cells expressing AdipoR1.

PAGE 14

14 CHAPTER 1 INTRODUCTION Iron and Zinc Homeostasis Zinc and iron are essential nutrients for all organism s and play important roles in many biochemical processes. Zinc, a strong lewis acid, is relatively inert. It is an essential component of over 300 enzymes, such as Cu/Zn dismutase a nd alcohol dehydrogenases, and plays a critical structural role for many prot eins (Vallee and Auld, 1999; Eide, 1998). Iron, unlike zinc, can donate and accept electrons and is an important cofactor in electron transport and many redoxactive metalloenzymes. Since iron and zinc are th e most abundant trace elements in human cells (Lyons et al., 2000), a deficiency or excess of thes e metals could lead to a variety of problems. Severe zinc deficiency is linke d to rheumatoid arthritis, centr al nervous system disorders, acodermatitis enteropathica, and impaired grow th and immune function (Brown et al., 1998) (Beyersman and Harase, 2001). Meanwhile, zinc t oxicity interferes with mitochondrial aconitase activity, thereby affecting cellular respiration (Coste llo et al., 1997). Iron de ficiency can lead to anemia, but an excess, is linked to hemochromatosis, which can l ead to organ failure and chronic diseases such as cirrhosis and diabetes ( Dallman et al., 1980). Becaus e of these considerations, organism s have evolved regulatory systems to cont rol uptake, distribution, and detoxification of metal ions. Most of these regulatory system s were first identified in the budding yeast, Saccharomyces cerevisiae, thus making it an excellent model system; first, metal transport has been extensively studied in this organism; second, S. cerevisiae has been used to identify genes involved in metal transport in higher eukar yotes, including humans (Ho et al., 2002). Zinc uptake in S. cerevisiae is controlled in response to zinc levels inside the cell. During zinc limitation, ZRT1 the high affinity zinc transporter, is induced (Zhao and Eide, 1996). Similarly low affinity zinc transporter is induced by low zinc levels and controlled by ZRT2

PAGE 15

15 transcription. Regulation of these genes is mediated by the ZAP1 ( Z inc-responsive A ctivator P rotein) gene (Figure 1-1), which encodes a tr anscriptional activator with seven carboxyterminal C2H2 zinc finger domains and two amino-term inal activation domains (Zhao and Eide 1997). Zap1p binds to a ZRE ( Z inc R esponse E lement) consensus sequence, 5ACCTTNAAGGT, in the promoters of ZRT1 ZRT2 ZRC1 ( Z inc R esistance C onferring, vacuolar membrane storage transporter), FET4 (low affinity zinc transporter) and ZAP1 itself (Figure 1-2) (Zhao et al .,1998; Lyons et al., 2000). The availability of iron in the cell affects the regulation of iron uptake in S. cerevisiae which is sensed by the transcription factor Af t1p. Aft1p is known to bind to a consensus sequence known as the FeRE ( Fe R esponse E lement) (Figure 1-3), 5-TGCACCC, in the promoter region of a variety of genes. During iron limitation, many genes are induced, including, FRE1 ( F erric RE ductase), FRE2, FET3 ( FE rrous T ransport), and FTR1 ( F e TR ansporter) (Babcock et al., 1997; Dancis et al., 1992). Unlike the zinc uptake system, the high-affinity iron uptake system is highly specific for Fe3+. This specificity is conferred by multiple chemical reactions in which Fe3+ is first reduced by Fre1p or Fre2p, then handed to Fet3p that oxidizes Fe2+ to Fe3+. Finally Ftr1p, which is direc tly coupled to Fet3p, allows Fe3+ to enter the cell (Askith and Kaplan, 1997). Figure 1-4 shows the membrane transport proteins involved in the acquisiti on of iron in yeast. Identification of the IZ H s In order to define the Zap1p regulon and iden tify genes involved in zinc metabolism, DNA microarrays and computer analysis of shared motifs in the promoters of co-regulated genes were performed by Lyons, et al (Lyons et al., 2000). Thes e analyses identified 46 genes regulated by Zap1p. Two of these genes identified, Y DR492w and YOL002c encode

PAGE 16

16 homologous proteins that are induced by zinc deficiency in a Zap1p dependent manner. This was due to the presence of a ZRE found in the promoter region of YDR492w and YOL002c A third protein, YOL101c was identified in a separate micr oarray experiment to be induced by excess zinc (3 mM). Also in this high zinc experiment YOL002c was induced. An additional gene, YLR023c related to YDR492w YOL002c and YOL101c was identified by BLAST and PSI-BLAST searches. Since three of these four ge nes in yeast respond to changes in zinc levels inside the yeast cell, they were renamed IZH s ( I mplicated in Z inc H omeostasis): YDR492w ( IZH1 ), YOL002c ( IZH2 ), YLR023c ( IZH3 ), and YOL101c ( IZH4 ). The IZH s have been implicated in other roles independent of Zap1p regulation. First, IZH4 was found to respond to excess Zn2+, Co2+, and Ni2+ or deficiency in Fe3+ limitation by Mga2p, a transcription factor that regulates hypoxia. Second, IZH2 is regulated by the fatty acid, mystiric acid (Lyons et al., 2004). Last, they have seen to be al so involved in the regulation of genes involved in iron-upta ke (Kupchak et al., 2007). Structural Features The IZH genes encode proteins that belong to a la rge fam ily of membrane proteins called PAQR (Progestin and AdipoQ Receptors), whic h consist of four yeast and eleven human proteins. Alignments of these pr oteins in figures 1-5 and 1-6 dem onstrate that these proteins can be classified into three distinct classes: 1) Class I: Four yeast protei ns: Izh1p, Izh2p, Izh3p, and Izh4p Four human proteins: AdipoR1 ( Adipo nectin R eceptor), AdipoR2, PAQR3, and PAQR4 2) Class II: mPR ( m embrane P rogestin R eceptor), PAQR6, mPR mPR and PAQR9 3) Class III: PAQR10 and PAQR11

PAGE 17

17 Further investigation of protein sequences showed the PAQR proteins within each class exhibited close sequence similarity. In Class I, the yeast Izh1p and Iz h2p share 39% amino acid identity, while the human PAQRs, AdipoR1 a nd AdipoR2 exhibit 68% identity. Sequence similarity was also seen in Class II, where mPR can be paired with mPR and mPR with PAQR6. Figure 1-6 also showed Class II I PAQR10 and PAQR11 were very similar. The PAQR protein family consists of at least seven transmembrane (TM) domains (Figure 1-7), but lacks significant amino acid similarity. When the PAQR family of proteins is aligned, they exhibit only three highly conserved re gions located on the intracellular side of the membrane. The first motif is located at the start of the TM1, Ex2-3Nx3H, but can be expanded to PxnGxnS/TxnEx2-3Nx3H if the Class I and Class II PAQRs are considered. Motif two consists of an Sx3H at the end of TM2 and an aspartic acid at the beginning of TM3. Motif 3 has an Hx3H sequence in the loop region between TM6 and TM 7, which can be expanded to PGxnPExnHx3H if only Class I and Class II are considered. Also, hydropathy plots demonstrate that Class II PAQR proteins contain an a dditional region of hydrophobic ami no acids, which can be thought of as an eighth TM domain (Figure 1-8). Even though the topology of the PAQR protei ns is uncertain, it has been suggested that the PAQR proteins all have an internal N-termi nus and external C-terminus, except for the Class II PAQR proteins, which would have an internal C-terminus. However, the only assessment that could be confirmed at this time is that two hu man adiponectin receptors (AdipoR1 and AdipoR2) (Yamauchi et al., 2003) and yeas t Izh2p (Kim et al., 2006) and Iz h4p (Kim et al., 2003) have an external C-terminus. Based on structural pred iction programs, the Izh proteins have been suggested to be localized to the plasma membrane, while the human PAQRs localization remains in question, either residing in the plasma membra ne or endoplasmic reticulum. The conserved

PAGE 18

18 motifs found on the intracel lular loops may indicate the PAQR proteins all signal in a similar manner. The extracellular loop regions exhibit no homology, but this may explain why the Izh and human PAQR proteins bind to three di fferent ligands: osmotin, adiponectin, and progesterone. Considering that the conserved regions are intr acellular, we propose that yeast IZH s or human PAQRs exhibit a conservation of signaling mechanism. The mPRs have often been confused with the G-protein coupled receptors (GPCRs) based on their topology and mechanis m of signal transduction. Firs t, they both are predicted to possess seven TM domains with an extracellular N-terminus and an intracellular C-terminus. Second, pertussis toxin, an inhibitor of Gi, also inhibits the effects of mPRs (Zhu et al., 2003a). However, these characteristics do not prove the PAQRs are GPCRs. To refute these claims, Gi involvement in GPCR-mediated sign aling would have to be direct but this may not be the case for the mPRs. Also the PAQRs and GPCRs sh are no sequence homology beyond the presence of seven predicted TM domains (Lyons et al., 2004 ); however, the topology and number of TM domains remains unconfirmed, since no groups have mapped the PAQR protein topology. Last, the topology of the GPCRs appears to opposite si de to that of AdipoR1 and AdipoR2 (Yamauchi et al., 2003). Thus the similarities between th e PAQR and GPCR proteins are only speculative, but this does not rule out the PAQR s are distantly related to GPCRs. Class I, II, and III Receptors The yeast Izh and human PAQR receptors can be classified into three categories based on sequence similarities. The human adiponec tin receptors, AdipoR1 (PAQR1) and AdipoR2 (PAQR2), and the osmotin receptors, OsmoR (Izh2p), found in various species of yeast make up the class I receptors. There also exists a similarity between the human adiponectin receptors and the yeast Izh1p and Izh2p. However, Izh3p a nd Izh4p along with PAQR3 and PAQR4 are more

PAGE 19

19 divergent and distantly related but st ill can be considered to be part of this class. Most of the class I proteins can be found throughout metazoan evolution ranging from fungi to humans. One of these receptors, Izh2p, was recently discov ered to bind to osmotin (Narasimhan et al., 2001). Osmotin is a 24 kDa protein that be longs to family known as PR-5 defensins. Moreover, osmotin was first identified in toba cco plants and exhibits antifungal properties against the pathogenic Candida albicans and the non-pathogenic Sacchromyces cerevisiae (Koiwa et al., 1997). Osmotin is not only induc ed by desiccation and salinity, but also by abscissic acid, wounding and cold temperatures (Zhu et al., 1993; Zhu et al., 1995). Osmotin is suggested to be involved in a pathway that requires the Ras2p G protein, which ultimately may lead to apoptosis of the yeast cell (Narasimhan et al., 2005). Early theories postulated that osmotin may induce holes in the fungal wall th ough a specific interaction with the plasma membrane (De Vos et al., 1985) (Roberts and Se litrennikoff, 1986). However, the mechanism by which osmotin starts a signaling cascade that leads to plasma membrane permeabilization remains unknown. The human Class I PAQRs, AdipoR1 and Adi poR2, bind to the adipose secreted ligand known as adiponectin (Yamauchi et al., 2003). Adiponectin is a 244 amino acid protein of approximately 30 kDa that is encoded by the ad iponectin gene, apM1, located on chromosome 3q27. Also called gelatin-binding protein-28 (GBP28), ACRP30 or AdipoQ, adiponectin belongs to the complement C1q protein family (S cherer et al., 1995) to which the inflammatory tumor necrosis factorbelongs. Originally, adiponectin was thought to be expressed and secreted only by adipocytes (Hu et al., 1996), but recent studies have de tected adiponectin in skeletal muscle and liver cells (Yamauch i et al., 2003; Dela igle et al., 2004).

PAGE 20

20 Adiponectin is one of the most abundant proteins found in human plasma, with concentrations ranging from 3 to 30 M (Chandran et al., 2003). Unlike many adipose derived proteins, levels of adiponectin ar e found to be lower in obese than in lean individuals. Studies have shown that low levels of ad iponectin are associated with di seases such as obesity, insulin resistance, and type 2 diabetes, cardiovascular disease and hypertension (Trijillo and Scherer; 2005). Additionally, adiponectin expression is down regulated by insulin and recent studies suggest adiponectin levels may be affected by dietary factors (Qi et al., 2005). In this case, it has been shown that calorie restriction and low glycem ic diets can raise adiponectin levels (Qi et al., 2005). Low levels of adiponectin have also been reported to be associated with certain types of malignancies, such as endometria l cancer (Petridou, et al., 2003), breast cancer (Mantzoros et al., 2004), leukemia (Petridou et al., 2006), colon caner (Ishikawa et al., 2005), and prostate cancer (Goktas et al., 2005). Adiponectin consists of four domains, including a signal peptid e at the N-terminus, short variable region, a collagen-like motif a nd a C-terminus globular domain. Additionally, adiponectin is known to circulat e in two forms. First, th e low molecular weight (LMW, globular) oligimer consists of two trimers a nd second, the high molecular weight (HMW, full length) oligimer is made of four to six trimers, which is the predominant form in human plasma (Pajvani et al., 2004) (Figure 1-9). The two forms of adiponectin possess distinct signaling activities. Both the HMW and LMW forms induce NFB activation, whereas the LMW form induces AMP-protein kinase (AMP K) activation in muscle (Tsao et al., 2002). The activation of AMPK increases glucose uptake (K urth-Kraczek et al., 1999), glycoge n formation (Merrill et al., 1997), and inhibition of acetyl-CoA carboxylase (ACC), which leads to decreased fatty acid synthesis (Tomas et al., 2002). Adiponectin ma y also have additional roles in controlling

PAGE 21

21 inflammation and having anti-atherogenic properties thought to be caused by monocytes adhering to the blood vessel walls (Ouchi et al., 1999). AdipoR1 and AdipoR2 are derived from di fferent genes located on chromosomes 1q23 and 12p13 that encode 376 and 386 amino acid polype ptides, respectively. Each receptor has been shown to have different dist ributions and affinities for the two forms of adiponectin. Also, AdipoR1 and AdipoR2 exhibit significant similarity in showing 67% amino acid identity (Yamauchi et al., 2003). AdipoR1 shows a high a ffinity for globular adiponectin, but can also bind to full length adiponectin at a lower affinity, and is expresse d mainly in skeletal muscle; however, it is also found in liver and endothelial cells (Goldstein and Scalia; 2004). Meanwhile, AdipoR2 can be found at high levels in the liver a nd exhibits moderate affi nity for globular and full length adiponectin (Golds tein and Scalia; 2004). The Class II receptors, a new class of membra ne progestin receptors (mPR) that mediate rapid, non-genomic steroid actions, were identifie d in spotted seatrout ovaries, and shown to mediate the induction of oocyte maturation (Zhu et al., 2003a). Furthermore, the Class II receptors are presumed to contain a hypothetical eighth transmembrane domain, which is a major difference between the Class I and Class III proteins. These receptors are known as mPR (PAQR7), mPR (PAQR8), mPR (PAQR5) and bind to progeste rone (Zhu et al., 2003b). Additionally, Class II also includes the proteins PAQR6 and PAQR9. Progesterone is a C-21 steroid hormone that is required for many aspects of female reproduction, including sexual behavior, gonadotrophi n secretion, implantation of blastocyst, and maintenance of pregnancy (Mulac-Jericevic et al., 2000) (Conneely et al ., 2002). This hormone is produced in the adrenal glands, ovaries, cor pus luteum after ovulati on, the brain, and in the

PAGE 22

22 placenta during pregnancy. Progesterone also has therapeutic applications ranging from birth control to the treatment of endometrial cancers (Sitruk-Ware and Plu-Bureau, 1999). Progesterone, like all other steroids, is s ynthesized from pregnenolone, a derivative of cholesterol. The conversion of pregnonolone to progesterone take s place in two steps. The 3hydroxyl group is converted to a keto group and a double bond is moved from C-5 to C-4. Also, progesterone is the precursor of the mineralocorticoid aldosterone and after conversion to 17 hydroxyprogesterone can be synthesized into cor tisol and androstenedi one. Androstenedione can be further converted to testos terone and estradiol (Figure 1-10). Many of the effects of pr ogesterone occur thr ough nuclear progesterone receptors; however, it has been known for the last 30 years that some of the effects occur too rapidly to be involved in the classical signali ng pathway (Edwards, 2005). This type of signal transduction involves the steroid entering the target cell by passive diffusion through the plasma membrane, passing through the nuclear membrane and binding to a nuclear receptor. The steroid-receptor complex then undergoes a conformational change and binds to DNA sequences known as steroid response elements. There the steroid-receptor acts as a transcription factor modulating gene transcription and synthesis of mRNAs and protei ns. However, there are examples where the action of steroids cannot be explained by the clas sical signaling model. Th is signaling occurs far too rapidly to require gene tran scription and may require the existence of integral membrane receptors that bind steroids at the cell surface and generate intracellular second messengers. This can be seen in highly specialized cells, such as spermatozoa, that are transcriptionally and translationally silent (Correia et al., 2007). This model has been called non-genomic signaling because it does not require transcription or tran slation to elicit its effect (Boonyaratanakornkit

PAGE 23

23 and Edwards, 2007). The mPRs are also though t to signal via non-genomic mechanism because of their abilty to generate the second messenger sphingoid bases. mPR mPR and mPR are derived from genes located on chromosomes 1p36, 6p12, and 15q23, respectively that encode proteins of 346, 359, and 330 amino acids. Even though all three isoforms bind to progesterone, they are expressed in different tissues. The -isoform is expressed in reproductive ti ssues, particularly in the placenta, testes, and ovaries, but also in the kidneys. The -isoform is found only in the brai n and the spinal cord, while the -isoform is present mainly in the kidneys and colon, but also found in the adrenal gland and lungs (Zhu et al., 2003a). Human PAQR10 and PAQR11 proteins make up the final class of receptors. Their function is thought to be invol ved in monocyte to macrophage differentiation (Brauer et al., 2004). These proteins are distantly related to class I and II PAQRs; however, they are related to proteins from the bacterium, Baciluus cereus, named hemolysin III (Hly-III) (Baida and Kuzmin, 1996). Hly-III are known to have hemolytic activity and form pores in the outer membrane of the host cell (Baida and Kuzmin, 1996), but it is not known if this action occurs by direct or indirect means. Thus, more research is needed in this area to identify the function of receptors in this class. Signaling of FET3 In our study, we will show a signaling mechanism in which increased PAQR gene dosage repressed FET3 (Figure 1-11). This task was perform ed using a bottom up approach, where we first identified stress res ponsive transcriptional f actors that are responsi ble for activation (Msn2p and Msn4p) and repression (Nrg1p and Nrg2p) of FET3 We will then identifiy several upstream signaling proteins required for this physiological effect. In particular, we will show the

PAGE 24

24 involvement of AMPK, PKA, and the sphingoid base-sensing Pkh1p/Pkh2p. Finally, we will demonstrate that PAQR overexpression increase s sphingoid bases, which, ultimately, leads to repression of FET3 NRG and MSN To survive in harsh conditions, m any organisms have adapted to changes, including nutrient deficiency, changes in osmolarity, pH, temperatur e, and oxidative stress. Saccharomyces cerevisiae adapt by expressing different sets of genes (Mager and De Kruijff, 1995). One set of transcription factors that responds to stresses are the partially redundant (C2H2) zinc finger transcription factors MSN2 ( M ultiS uppressor of SN F1 mutation) and MSN4 (Martinez-Pastor, et al., 1996). During times of stress such as heat shock, osmotic stress, and nutrient depletion, these transcription factor s localize to the nucleus and activate stress responsive genes by binding to th e stress response elements (STR E; CCCCT) in their promoters (Martinez-Pastor et al., 1996). Another set of tran scription factors that regulate stress response genes are Nrg1p (N egative R egulator of G lucose-repressed genes) and Nrg2p. However, these proteins function as transcrip tional repressors (Park et al., 1999). Nrg1p and Nrg2p bind to the NRE ( N RG R esponse E lement; CCCTC) and are known to repress FLO11 and GAL genes, which are implicated in glucose metabolism (Kuc hin, et al., 2002) (Park et al, 1999). These transcription factors also nega tively regulate haploid invasive growth as well as diploid pseudohyphal growth (Kuchin et al., 2002). Protein kinase A (PKA) is known to be upstr eam of the stress response activators Msn2p and Msn4p, where PKAs signal causes the stress response to be down-regulated (Boy-Marcotte et al., 1998). This occurs by PKA phosphorylat ing Msn2p, which causes it to accumulate in cytoplasm and stop the stress res ponse (Smith et al, 1998).

PAGE 25

25 Upstream of the Nrg1p and Nrg2p proteins is Snf1p, the catalytic subunit of the AMPdependent kinase complex. The binding si tes for Nrg1p/Nrg2p and Msn2p/Msn4p are very similar. Recent studies suggest these two sets of transcription factors co mpete for binding sites on a subset of stress response pr omoters (Vyas et al., 2005). The CCCTC site can also be found in the promoter regions of both FET3 and ZRT1 which may implicate these transcription factors in the regulation of iron a nd zinc homeostasis. AMPK The AMP k inase (AMPK) complex is a hetero trimer composed of a catalytic -subunit, -subunit, and a regulatory -subunit (Figure 1-12). In Saccharomyces cerevisiae the SNF1 ( S ucrose N onF ermenting) gene encodes, the catalytic -subunit and is a seri ne/threonine protein kinase that is activated during ti mes of glucose limitation (Woods et al, 1994). It is required for expression of glucose repressed ge nes involved in utilization of ga lactose, sucrose, raffinose, or nonfermentable carbon sources, as well as resp iration, gluconeogenesi s, filamentation and sporulation (Celenza and Carlson, 1986) (ThompsonJaeger et al, 1991). Snf1p has been found to interact with other proteins including the -subunit Snf4p. Snf4p is also required for transcription of glucose repressed genes and func tions as a Snf1p activat or (Celenza and Carlson, 1989). The -subunits interact with both Snf1p and Sn f4p, and act as a scaffold to hold the heterotrimer complex together. S. cerevisiae express three -subunits encoded by the GAL83 ( GAL actose metabolism), SIP1 ( S NF1 I nteracting P rotein), and SIP2 genes. At least one of the -subunits needs to be expressed with each subun it conferring functional specificity to Snf1p (Schmidt and McCartney, 2000). However, the major role of the -subunits is to control localization of the AMPK comple x (Vincent et al, 2001). During growth in high glucose media, all three of the -subunits are localized to the cytoplasm, but when yeast are switched to nonfermentable carbon sources, Gal83p-Snf1p moves to the nucleus, Sip1p-Snf1p shifts to the

PAGE 26

26 vacuole membrane, and Sip2p-Snf1p remains in the cytoplasm. Activation of AMPK occurs by three upstream kinases, known as Sak1p ( S nf1p A ctivating K inase), Tos3p ( T arget O f S BF), and Elm1p ( EL ongated M orphology). PKA Protein kinase A (PKA, cAMP-dependent kina se ) plays a major role in metabolism and cell proliferation based on nutrient availability. For PKA ac tivation, a high level of cAMP is required and occurs by Ras2p through the activati on of adenylate cyclase. PKA is activated by cAMP binding to the regu latory subunit Bcy1p ( B ypass of CY clase mutations), which allows the release of three cataly tic subunits encoded by TPK1 ( T akashis P rotein K inase), TPK2 and TPK3 genes (Thevelein, 1994). Furthermore, PKA regulates genes involved in iron uptake and respiration; however, each catalytic subunit has its own function. Tpk2p represses genes involved in high-affinity ir on uptake but induces pseudohyphal growth. Tpk1p derepresses genes involved in branched chain amino acid formation and also has a role in controlling in iron levels (Robertson et al., 2000). PDK ( PKH1 and PKH 2 ) The next step was to find what enzymes act upstream of PKA and AMPK. A clue came from trying to elucidate the IZH genes, where IZH1 and IZH4 were found to be divergently transcribed from a pair of kinases, PKH1 and PKH2 (Figure 1-13). In yeast, genes that share the same promoter are often co-regulated and belong to the same chemical pathway (Kruglyak and Tang, 2000) Pkh1p ( P kb-activating K inase H omolog) and Pkh2p are partially redundant isoforms that are homologs of human PDK1 ( P hosphoinositide D ependent K inase) (Casamayor et al, 1999). PDK1 is a member of the AGC family of kina ses, that is a main regulator of downstream

PAGE 27

27 kinases, such as PKB ( P rotein K inase B ), S6K (p70 ribosomal S6 K inase), and SGK ( S erumand G lucocorticoidstimulated protein K inase) (Vanhaesebroeck and Alessi, 2000). But how are Pkh1p and Pkh2p activated? In hu mans, PDK is activated by four signals: autophosphorylation, 3-phosphoinositides, TOR2 and sphingosine (Kim et al., 2002). But in yeast, the lipid (phyto)sphingosine is the only known activator of both Pkh1p and Pkh2p (Liu et al., 2000). Sphingolipids Sphingolipids are one of the main lipids comp rising membranes in eukaryotic cells. But sphingolipids also play a role as second messenge rs in regulating signal transduction pathways, mediating cell growth, differentiation, and in cell death (Futerman and Riezman, 2005). Importantly, the ratio of sphingos ine to ceramide inside the cell may act as a rheostat where sphingosine controls cell prolif eration and ceramide has roles in apoptosis (Pyne and Pyne, 2000). To understand specific functions of sphingo lipids and their derivati ves, especially the identification of genes involve d in sphingolipid metabolism, many researchers are using the simple model organism, Saccharomyces cerevisiae (Dickson and Lester, 2002) (Sims et al, 2004). Structurally, sphingolipids are composed of a long chain base (LCB), a fatty acid, and a polar head group. In yeast, two kinds of LC Bs can be found, dihydrosphingosine (DHS) and its 4-hydroxy derivative, phytosphingosine (PHS). The first and rate limiting step in sphingolipid synthesis (Figure 1-14) is the formation of 3-ketodihydrosphingosine from pa lmitoyl-CoA and serine. This reaction, which occurs on the cytoplasmic surface of the endoplasmic reticulum (E R) (Mandon et al., 1992), is catalyzed by the two subunit (Lcb1p (L ong C hain B ase) and Lcb2p) serine palmit oyltransferase (Buede et al., 1991; Pinto et al., 1992) Y east cells defective in LCB1 and LCB2 genes are lethal and cannot

PAGE 28

28 grow without LCB supplementation (Wells et al., 1998). But in humans a mutation in LCB1 causes sensory neuropathy, a disease that causes degeneration of the dorsal root ganglia and motor neurons (Gable et al., 2002). Tsc 10p catalyzes the second step in which 3ketodihydrosphingosine is reduced to DHS (Beeler et al., 1998). Then, two ceramide synthases, Lag1p ( L ongevity A ssurance G ene) and Lac1p ( L ongevity A ssurance gene C ognate) (Guillas et al., 2001), react with DHS and a C26 fatty acid Co A to make dihydroceramide. Dihyroceramide can then be hydroxylated at C4 by Sur2p ( SU ppressor of R vs161 and rvs167 mutations) to yield phytoceramide. However, phytoceramide can also be formed from DHS by Sur2p to form PHS, then amide linked to the C26 fatty acid. Furthe r modification can occur wh en yeast are grown in the presence of oxygen, where the enzyme Scs7p hydroxylate the fatty acid group at C2 and C3 positions (Haak et al., 1997). An important aspect of the ceramide synthesi s is the formation of C26 fatty acids from C14-C18 fatty acids. These reactions take place in the ER and require several genes, including ELO1 ( ELO ngation defective (Toke 1996), FEN1 ( FEN propimorph resistance), SUR4 (Oh 1997), TSC13 ( T emperature-S ensitive C sg2 suppressor) (Kohlwein et al., 2001), YBR159w (Han 2002 JBC), and ACP1 ( A cyl C arrier P rotein) (Gaigg et al., 2001). Once C14 and C16 fatty acid are formed from acyl-CoA, Elo1p starts enlongati ng the fatty acid chain. Fen1p then performs the condensation reaction with C20 and C22 fatty acyl-CoAs, while Sur4p does the same for C22 and C24 fatty acyl-CoAs (Rossler et al., 2003). Ybr159w is proposed to reduce 3-ketoacyl-CoA to 3-hydroxyacyl-CoA for fatty acids greater than 18 carbons (Han et al., 2002); however, the protein that is involved in dehydrating 3-hydroxyacyl-CoA to the enol intermediate is unknown. The next step is the elongation of the fatty acid catalyzed by Tsc13p, which reduces the enol.

PAGE 29

29 Finally, Acp1p is needed in this process to help deliver the fatty acid CoA to start the process of elongation. Once ceramides are synthesized, they are transported from the ER to the Golgi apparatus where the polar head substituents can be added at C1. In human cells there are hundreds of sphingolipid molecules that differ in their head gr oup. But in yeast there ar e only three classes of sphingolipids. The simplest y east sphingolipid, inos itolphosphorylceramide (IPC), is made by transferring inositolphosphate from phosphatidate to ceramide by the enzyme Aur1p ( AU reobasidin A R esistance) (Hashida-Okado et al., 1996) Subsequent maturation of IPC requires Sur1, Csg2 ( C a2+ S ensitive G rowth), and Csh1 ( C SG1 / S UR1 H omolog), which adds mannose to form mannosyl-inositolphosphorylcerami de (Zhao et al., 1994). Finally, MIPC is converted to mannosyl-diinosit olphosphorylceramide (M(IP)2C) by Ipt1p (Dickson et al., 1997). Once formed, the three classes of sphingolipids can localize to plasma membrane where they make up 30% of all the lipids (Patton and Lester, 1991). The breakdown of sphingolipids, especially in mammals, is essential for survival. Many diseases occur due to lack of turnover of one or more sphingolipids (R aas-Rothschild et al., 2004). Breakdown begins with Isc1p ( I nositol phospho S phingolipid phospholipase C ) hydrolyzing IPC, MIPC, and M(IP)2Cs polar head group to yield ceramide (Ella et al., 1997) (Sawai et al., 2000). Isc1p in S. cerevisiae plays a role in stress tolerance as sodium chloride and lithium chloride have been shown to cause growth defects in an isc1 -deleted strain (Betz et al., 2002). Ceramides, whose concentration levels are controlled by de novo synthesis, have many roles ranging from apoptosis to stress re sponses (Ogretmen and Hannun, 2004). However the inability to deacylate the am ide bond between sphingosine and fatty acid in humans leads to

PAGE 30

30 lysosomal storage condition known as Farbers di sease (Koch et al., 1996). In yeast, Ypc1p ( Y east P hyto C eramidase) and Ydc1p ( Y east D ihydroC eramidase) are the ceramidases that yield either PHS or DHS (Mao et al., 2000). PHS or DHS derived from either de novo synthesis or ceramide breakdown can be phosphorylated by two LCB kinases, Lcb4p and Lc b5p (Nagiec et al., 1998), to give PHS-1phosphate (PHSP) and DHS-1-phosphate (DHSP). Two fates arise for PHSP and DHSP. One, they can be catabolized by Dpl1p (Saba, et al ., 1997) to yield ethalomine-phosphate and fatty aldehydes. Two, they can be dephosphorylated by Lcb3p and Ysr3p (Y east S phingolipid R esistance) (Qie et al., 1997; Mao et al., 1997) to again form PHS and DHS. Functionally, PHSP and DHSP are recognized as signaling mol ecules and mediate processes such as cell growth and survival and calcium mobilization (Pyne and Pyne, 2000) (Spiegel and Milstien 2003). Yeast as a Model Organism Saccharomyces cer evisiae, commonly known as bakers, brewers, or budding yeast, has been a standard laboratory model since the 1950s Yeast has many attributes in biological studies such as being a hearty organism, exhibiting rapid growth, having the ease of replica plating and mutant isolation, a well-defined genetic system, and being a versatile DNA transformation system (Gietz and Woods, 2002). Additionally, yeas t is a non-pathogenic organism and can be handle d with little precaution. In contrast to other organisms, Saccharomyces cerevisiae can exist in either a haploid or diploid state, and are viable with a large number of markers. The development of DNA transformation has made yeast an ideal organi sm for gene cloning and genetic engineering techniques. Structural genes co rresponding to almost any genetic trait can be identified from complementation from plasmid libraries. Gene s can be expressed in yeast cells through

PAGE 31

31 replicating plasmids or by integration of th e gene into the genome. Unlike many other organisms, recombination of transforming DNA in yeast occurs by homologous recombination. Thus, cloned yeast sequences, along with foreig n sequences on plasmids, can be directed to specific location in the genome. This allows the user to replace a certai n gene with its mutated counterpart, which enables the researcher to make a mutant library. In fact, researchers can use an abundance of tools that are co mmercially available, such as a full set of deletion mutants and tagged open reading frames to locali ze or purify the protein of choice. In using yeast, several genetic databases exis t, which is an important tool for developing basic knowledge about the function and organization of eukaryotic cell genetics and physiology. Also, a large group of yeast rese archers contribute a wealth of information and expertise. Saccharomyces cerevisiae was the first eukaryote whose genome was completely sequenced (Goffeau, et al., 2006). Thus, yeast has become one of the most important organisms for genomic research (Kumar and Snyder, 2001) and has been useful for studying proteinprotein interaction by two hybrid analysis (Uetz et al., 2000), protein localization (RossMacDonald et al., 1999), enzymatic activities (Martzen et al., 1999), genome-wide analysis of gene function by gene disruption (Oliver, 1996), and serial analysis of gene expression (Velculescu et al., 1997). One advantage of using yeast is that mamma lian genes can be introduced into yeast for functional analyses of the corr esponding gene products. At this time, there have been 6,000 genes recognized in yeast while, over 25,000 have been discovered in humans (Goffeau et al., 2006). Intriguingly, one-third of yeast proteins share homology to human proteins and 34% of disease-related human genes have functional homologs in yeast (Sturgeon et al., 1996). This

PAGE 32

32 high conservation of metabolic and regulatory m echanisms has contributed to yeast becoming the model eukaryotic system fo r diveresed biological studies. Additionally, yeast has been used to elucidate a variety of basic biological processes due to their simplicity of experimentation. For metal ion homeostasis, organisms have evolved complex mechanisms that have been conser ved throughout evolution. Multiple transport systems exist for these metals allowing organisms to obtain these metals under varying conditions. Yeast is an ideal model system because many of the individual proteins and pathways are conserved from yeast to huma ns (Valentine and Gralla, 1997). Thus, the information gained can have clinical applica tions such as improving dietary zinc and iron availability in humans and treating metal-related disorders (De Freitas et al., 2003). Summation In our initial studies, we exam ined several y east proteins that were implicated in zinc homeostasis; however, as we studi ed these proteins more closely we saw they also acted in iron homeostasis, as well as lipid metabolism and medi ating signaling events. We also noticed that the yeast proteins were related to eleven human proteins that bind to ligands such as adiponectin and progestin. We hypothesize that overexpre ssion of the protein or ligand binding to its receptors will hydrolyze phytoceramide to phytoc eramide, which in turn will activate Pkh1p/Pkh2p, protein kinase A and deactivate AMPK that will result in repression of the high affinity iron and zinc uptake systems by activ ating the repressors, Nrg1p/Nrg2p, and deactivating the activators, Msn2p/Msn4p (Figure 1-11). The wo rk outlined in this study will focus on the proteins involved in the high affinity iron and zinc uptake system, as well as the proteins participating in this signaling pathway. The focu s of this work entails studying ligand binding to its receptors. We also studied exogenous and endogenous sphingoid lipids and their effects on these uptake systems. By investigating these membrane protein receptors in yeast, we may be

PAGE 33

33 able to understand their effect on metal homeostasis and may one day apply this information to mammalian systems.

PAGE 34

34 Zap1p-Target Genes Figure 1-1. The transcription factor Zap1p binds to the zinc response element (ZRE) on the promoter region of genes in Saccharomyces cerevisiae during periods of low zinc. Zrt1pZn2+Zrt2p Vacuole Zrc1p Zrt3p Nucleus Fet4p Zn2+Zn2+ Figure 1-2. A Saccharomyces cerevisiae cell showing the zinc transporters.

PAGE 35

35 Aft1p-Target Genes Figure 1-3. The transcription f actor Aft1p binds to the iron response element (FeRE) on the promoter region of genes in Saccharomyces cerevisiae during periods of low iron. Fre1pFe2+Fe2+Fe2+Fe3+Fe3+Fe3+ Fre2p Fet3p Ftr1p FeRE Aft1p FET3, FTR1 FRE1, FRE2 Figure 1-4. A Saccharomyces cerevisiae cell showing the iron transporters.

PAGE 36

36 Figure 1-5. Multiple sequence alignment of PA QR proteins. Areas highlighted are highly conserved motifs.

PAGE 37

37 Figure 1-6. Phylogene tic tree displaying IZH and human PAQR prot eins. A bootstrapped phylogentic tree showing the relationship be tween PAQRs, alkaline ceramidases and GPCRs. Yeast Izh proteins cluster with human adiponectin receptors. The GPCRs clade is included as an outgroup to root th e phylogentic tree. The length of the tree branches is proportional to th e calculated distance between sequences with the scale bar indicating 0.1 substitutions per site. Nu mber at the nodes are confidence values that refer to the number of times per 1,000 tr ees drawn a particular grouping is made.

PAGE 38

38 Figure 1-7. Topology predicted for all members of the PAQRs with the locations and the consensus sequences of the three highly conserved motifs. Grey represents membrane, ovals indicate transmebrane domains, and circles represent conserved regions of PAQR family. Class I an d III PAQRs are presumed to have 7 transmembrane domains while Class II PAQRs may have 8 transmembrane domains.

PAGE 39

39 Figure 1-8. Kyte-doolittle plot of PAQR and PAQR-like prot eins. (A) Class I PAQR vertebrate homologs of AdipoR1. (B) Class II PAQR vertebrate homologs of mPR (C) Class III PAQR homologs of PAQR11. (D) Alkaline ceramidases.

PAGE 40

40 B.Monomer TrimerHexamer HMW Multimer LMW Figure 1-9. Adiponectin domains and forms. (A) Human adiponectin domain structure. Numbers indicate amino acids in each region. (B) Multimer formation of adiponectin.

PAGE 41

41 O O CH3CH3H HH CH3H O O CH3HH CH3H OH OH O CH3HH CH3H OH O O CH3CH3OH HH CH3H O CH3CH3H HH CH3H OH O O CH3HH CH3H H OH CH3HH H OH OH CH3HH CH3H O OH O CH3CH3OH HH CH3H OH O O CH3HH CH3H OH H OH O O CH3HH CH3H OH OH OH O O HH CH3H OH O H OH O CH3HH CH3H H CH3CH3CH3 Cholesterol Pregnenolone Progesterone Aldosterone Testosterone Cortisol 17 -Estradiol Corticosterone 17 -Hydroxyprogesterone 17 ,21-hydroxyprogesterone 21-Hydroxyprogesterone Dehydroisoandrosterone 17 -Hydroxypregnenolone Figure 1-10. Synthesis of progestins, glucoc orticoids, mineralocorticoids, androgens, and estrogens. All steroids synthesized are from pre gnenolone via cholesterol.

PAGE 42

42 inactive (cyto)Nrg1p/Nrg2p Nrg1p/Nrg2p active inactive AMPK AMPKactiveMsn2p/Msn4p Msn2p/Msn4pactive active active inactivePKA PKA active inactivePkh1p/Pkh2pPkh1p/Pkh2p ?Phytoceramide Phytosphingosine Ligand (cyto) (nucl) (nucl) (nucl) (cyto) (cyto) (nucl) CCCTC FET3 Figure 1-11. A model for the mechanism of FET3 by PAQR overexpression. Solid arrows indicate protein activation or localization of protein. Br oken arrows indicate protein interaction on localization a nd activation. Cyto = protein is cytoplamic. Nucl = protein is nuclear.

PAGE 43

43 Figure 1-12. The AMPK complex and activators. The AMPK complex is heteorotrimer composed of an -subunit (Snf1p), -subunit (Snf4p), and one of three -subunits (Gal83p, Sip1p, and Sip2p). Snf1p is activ ated by one of three AMPK kinases (Elm1p, Tos3p, and Sak1p). Figure 1-13. Saccharomyces cerevisiae genes IZH1 and IZH4 are divergently transcribed from PKH1 and PKH2

PAGE 44

44 CH3NH2OH OH OH CH3NH2OH OH CH3NH2OH O CH3NH OH OH OH R O O O OH OH N N NN NH2P O-O O P O O-O CH3CH3OH O NH O NH S O CH3 O NH2O HO H + LCB1, LCB2 TSC10 LIP1, LAG1, LAC1 YPC1, YDC1 SUR2 SCS7 Hydroxyphytoceramide Inosito-P-ceramideInositol Phosphate MIPC M(IP)2CInositol Phosphate GDP-mannose IPT1 SUR1, CSG2 AUR1 Phytoceramide Phytosphingosine Dihydrosphingosine 3-Ketodihydrosphingosine Palmityl-CoA L-Serine ATP ADPYSR3, LCB3 LCB4, LCB5CH3NH2O OH OH P OH OH O P O OH OH O NH2 O CH3 + Dihydrosphingosine-1-phosphate DPL1 Palmitaldehyde Phosphoryl-ethanolamine TSC3 Figure 1-14. Sphingolipid biosynthesis in Saccharomyces cerevisiae Not shown, phytosphingosine like dihydrosphingosine can be degraded in a similar manner to palmitaldehyde and phosphoryl-ethanolamine.

PAGE 45

45 CHAPTER 2 THE EFFECT OF IZH2 ON FET3 AND ZRT1 Introduction Iron and zinc are essential nutrients and play im portant roles in many biochemical processes. Zinc is a catalytic component of over 300 enzymes and has structural roles in many proteins, while iron can donate and accept elec trons and is a cofactor in many redox-active metalloenzymes (Eide, 2001). However, when thes e metals become scarce, an organism must find ways to increase uptake of iron and zinc to maintain growth. Therefore, organisms have evolved homeostatic regulatory systems to control uptake during nutrient depletion (Rutherford and Bird, 2004). In Saccharomyces cerevisiae, ironand zinc-acq uisition during periods of limitation are controlled by the transcription factors, Aft1p and Zap1p. Aft1p and Zap1p sense iron and zinc in response to deficiency, which induc e genes involved in uptake. However, recent studies have showed iron and zinc uptake systems may be induced during iron and zi nc sufficiency when Aft1p and Zap1p are turned off. The high affinity ironand zinc-uptake systems are also regulated by carbon starvation (Hau rie et al., 2003) and during diffe rent periods of the cell cycle (Cho et al., 1998). During severe iron limitation, Saccharomyces cerevisiae induces genes involved in ironacquisition by Aft1p. Iron is taken into the cell by the high affinity transport system (Stearman et al., 1996). Since iron in the e nvironment is found as insoluble Fe3+, it must be reduced to Fe2+, which occurs by two flavocytochromes containi ng ferrireductases, Fre1p and Fre2p (Georgatsou and Alexandraki, 1994). Fe2+ is then handed over to Fet3p, a ferroxidase, which reoxidizes the iron. Finally, Ftr1p transports Fe3+ across the cell membrane and into the cell (Askwith et al., 1994). Zinc uptake in S. cerevisiae is controlled by Zap1p in respons e to zinc levels inside the

PAGE 46

46 cell. During zinc limitation, ZRT1 the high affinity zinc transpor ter is induced (Zhao and Eide, 1996), while the low affinity zinc transporter is induced by low zinc levels and controlled by ZRT2 transcription. Recent studies have identified four proteins that have been implicated in metal ion homeostasis and named IZH1-4 ( I mplicated in Z inc H omeostasis) (Lyons et al., 2004). IZH2 functions as a receptor for the plant defensin, osmo tin, but it is still not kno wn if this represents its physiological role (Naras imham et al., 2001). Also, IZH1 and IZH2 are confirmed targets of Zap1p, a transcription factor that induces genes during zinc limitation. Additionally, IZH2 and IZH4 are positively regulated by excess metals via Mg a2p (Lyons et al., 2004). The Izh proteins belong to a larger family of membrane receptor proteins known as the P rogestin and A dipoQ R eceptors (PAQRs). In this study, we present data showing that overexpression of the IZH2 gene resulted in repression of a Zap1p-responsive re porter. For this reporter, a Z inc R esponsive E lement (ZRE) (Zap1p binding site located in the promoter) fr om the ZRT1 promoter was inserted into a minimal CYC1 reporter. This reporter responds to zi nc deficiency, but overexpression of IZH2 repressed this reporter. Control experi ments demonstrated that the effect of IZH2 overexpression on this reporter was not due to a generalized def ect in transcription, tr anslation, or aberrant trafficking of membrane proteins due to thei r overexpression. Additionally, we did not see an increase in zinc levels inside the cell, which suggests this eff ect was due to a Zap1p independent mechanism (Lyons et al., 2004). We also showed that IZH -dependent repression of the reporter may be due to a regulatory element, CCCTC, which overlaps the inserted ZRE. Furthermore, we demonstrate the full length ZRT1 reporter is also affected by IZH2 -dependent repression. Since IZH2 overexpression affects

PAGE 47

47 genes involved in high affinity zinc uptake, we decided to examine their effect on genes involved in high affinity iron uptake. We present data that IZH2 overexpression represses FET3 during iron limitation. Furthermore, the CCCTC motif, wh ich is a known NRG bindi ng site (Park et al., 1999), is also found in the FET3 promoter (Kupchak et al., 2007). We suggest that repression of FET3 occurs by activation of the repressors Nr g1p/Nrg2p and deactivation of the activators Msn2p/Msn4p (Figure 2-1), which may compet e for binding to the CCCTC site. This study suggests that IZH2 overexpression affects high affinity iron and zinc transporters during ironand zi nc-limitation. Also, this repre ssion may be due to a regulatory element, CCCTC, found inside the ZRE and FET3 promoter. Materials and Methods Yeast Strains and Plasmids The wild type strain (DY1457) was a gift of Dr. David Eide (University of W isconsin, Madison). The BY4742 wild type strain and the msn2 msn4 nrg1 and nrg2 mutant strains were purchased from Euroscarf ( http://web.uni-frankfur t.de/fb15/mikro/euroscarf/col_index.htm l ), while the wild type strain MCY5326 and msn2 msn4 nrg1 nrg2 and nrg1 nrg2 msn2 msn4 mutant strains were provided by Dr. Marian Carlson at Columbia University (Vyas, et al., 2005) The genotype of all strains are listed in Supplementary Table 1. The p FET3-lacZ plasmid (Gift of D. Eide) ha s the promoter region of the FET3 gene fused to the lacZ open reading frame and is URA3 or LEU2 selectable. The pFLO11-lacZ strain, in which the lacZ ORF has replaced the FLO11 ORF in the genome 1278b, was provided by Dr. Florian Bauer at the University of Stellenbosch, Ma tieland, South Africa (van Dyk, 2005). The p FET3lacZ plasmid has the promoter region of the p FET3 gene fused to the lacZ open reading frame and is URA3 selectable. p CYC-STRE (pCZ-oligo31/32, provided by Janet Treger, UCLA) contains a fragment of the DDR2 promoter that includes tandem st ress r esponse e lements

PAGE 48

48 (STRE) to which the Msn2p and Msn4p transcriptional factors bi nd (Treger et al., 1998). IZH2 was cloned from yeast genomic DNA into pRS316GAL1 ( LEU2 ) via homologous recombination as previously described (Lyons et al., 2004). The primers used to clone IZH2 are listed in Supplementary Table 2. Plasmids containing the GAL1 -driven TAP (Tandem Affinity Purification) tagged NRG1 and NRG2 constructs were purchased from Open Biosystems. All clones were sequenced to ensure that no errors occurred in PCR. lacZ Reporter Constructs The p ZRT1361 and pZRT1305 insertional plasm ids (Gift of D. Eide) have the lacZ gene driven by the various truncations of the ZRT1 promoter (-361 to +3 and 305 to +3). These plasmids were transformed into wild type (BY4742 and DY1457) backgrounds to obtain reporter strains. pFET3-lacZ episomal reporter plasmids, such as p FET3398, pFET3297, pFET3263, pFET3233 and pFET3197 have a lacZ gene that is driven by di fferent truncations of the FET3 promoter (-398 to +3,-297 to +3, -263 to +3, -233 to +3, and -197 to +3, respectively (Gift of A. Dancis, University of Pennsylvania)). Several di fferent reporters have be en obtained that have the lacZ gene driven by a minimal CYC1 promoter in which the native Upstream Activating Sequence has been replaced with fragme nts from various promoters. The p ZRT1ZRE construct (pDg2) contains ZRE1 from the ZRT1 promoter along with 8 base pairs of upstream and 10 base pairs of downstream flanking sequence (Gift of D. Eide). The construction of p IZH1-ZRE contains a similarly sized fragment containing the ZRE from the IZH1 promoter (pDgIZH1ZRE). (The exact sequences of the insert ed fragment are shown in figure 2-1C. p CYC1FeRE (pFL-W) contains the following fragment of the FET3 promoter containing the Fe3+ Response Element (FeRE, underlined): CTTCAAAAGTGCACCC ATTTGCAGGTGC.

PAGE 49

49 Galactosidase and Ferroxidase Assays Yeast transform ations were performed by the li tium acetate method to introduce plasmids into appropriate strains (Gietz and Woods, 2002). Yeast was grown for 24 hours in low zinc media (LZM) or low iron media (LIM) supplemented with varying concentrations of ZnCl2 or FeCl3. LZM or LIM is made zinc or iron deficient by 1 mM EDTA and is buffered at pH 4.0 with 20 mM sodium citrate. Galactose or raffinose (A cros #17629-30-0) was used as the carbon source. LZM or LIM contains normal concentr ations of minerals except for ZnCl2 or FeCl3, which are varied from 1 M to 1 mM. Cells grew to mid-log phase in LIM or LZM, where -galactosidase (Lyons et al., 2004) or ferroxidase (De Silva et al., 1995) assays were performed following published procedures. -Galactosidase assays measured the hydrolysis of 1-nitrophenyl-Dgalactopyranoside (ACROS Organics #128820050) to 2-nitrophenol and monitored at 420 nm, which was measured at 20 minutes and quenched by sodium bicarbonate. Galactosidase activity is measured in Miller units and was calculated as follows: (A420 x 1,000) / min x mL of culture x A600). Ferroxidase assays fo llowed the disappearance of Fe2+, from ferrous ammonium sulfate, which was measured at 5 minutes and quenched by ferrozine (ACROS Organics #410570050). The cells were centrifuged and the abso rbance of the supernatant was monitored at 564 nm ( = 0.0279 M-1cm-1). The rate of Fe2+ oxidation in the absence of cells was subtracted to give the final value. Ferroxidase assay is measured in mol Fe2+ oxidized per minute and was calculated as follows: (OD564 x 1 x 10-3 L) / (0.0279 M-1cm-1 x min x OD600 x 1.5 x 108 cells). Galactosidase and ferroxidase activitie s were presented as % of Control, which signifies activity expressed as percent of activity measured in the wild type strain carrying the empty vector control. All experiments include triplicate data points and each experiment was performed three times. Representative experi ments are shown. All error bars represent standard deviation within a single experiment.

PAGE 50

50 Ferrozine Plates Iron-lim ited plates were made by adding 20 M FeCl3 to plates containing synthetic media, galactose, and 1 mM of the ir on-specific chelator, ferrozine. Protein Isolation Frozen pellets were resuspended in 4 m L MI B buffer (0.6 M mannitol, 20 mM HEPES, 1 mM EDTA) to which the following inhibitors we re added immediately: 1 mM PMSF and 10 g/mL protease inhibitor cockta il (for fungal and yeast extracts Si gma #P8215). Cells were transferred to low binding microcentrifuge tubes and disrupted by adding acid washed beads and vortexing five times for 1 minute, resting 1 minute on ic e between each disrupti on. Cells were then centrifuged at 3,000 rpm at 4C for ten minutes and the supernatant was collected. The supernatant was spun at 45,000 rpm for 90 minutes at 4C. The pellet was collected and suspended in MIB buffer with 33% glycerol and frozen at -80C. Gel Electrophoresis Sa mples were diluted to 1.0 mg/mL and comb ined with equal amounts of 2x SBB buffer (3% bromophenol blue, 0.2 M Tris-H Cl pH 6.8, 20% glycerol, 10% -mercaptoethanol, 2% SDS). Equal amounts of total protein were load ed on a 10% SDS-PAGE gel and run at 120 volts for two hours. The gel was stained with Coom assie Blue stain (0.02% Coomassie Blue R-250, 5% glacial acetic acid, 40% methanol) for 1 hour and counterstained with De-stain solution (10% glacial acetic acid, 5% methanol) overnight. Western Blot Proteins were transferred onto Hybond P (PVDP) membrane (American Biosciences #RPN303F) with electrophoresis transfer for one hour in 10% methanol, 1x Tris-Glycine, 0.1% SDS. The membrane was washed twice with 1x TBST (0.1% Tween 20) and then blocked with 5% milk in 1x TBST for one hour at room temper ature. The membrane was incubated overnight

PAGE 51

51 at 4C with the primary antibody (HA-probe rabbit polyclonal antibody (Open Biosystems #sc805), 1:500 dilution) in 1% bovine serum albumin (BSA) in 1x TBST, followed by washing the membrane three times with 1x TBST at room temperature. The secondary antibody (Goat anti-rabbit HRP conjugate (Open Biosystems #sc2004), 1:5,000 dilution) in 1% BSA in 1x TBST was added for one hour at room temper ature followed by washing the membrane three times with 1x TBST at room temperature. Th e bands were detected with ECL Plus Western Blotting detection System (GE Healthcare #RPN2132). Results Effect of IZH2 Overexpression on the ZRT1 Promter Previous studies showed that the p ZRT1 -ZRE1 (also called pDg2) and pIZH1 -ZRE reporters respond to the amount of zinc added to LZM. -Galactosidase activity is fully induced when grown in LZM containing 1 M zinc and repressed in LZM containing 1 mM zinc. This occurs due to reporters cont aining a ZRE, to which Zap1p mu st bind for transcriptional activation. Overexpression of the IZH2 gene during zinc-deficiency results in repression of pZRT1 -ZRE. But the overexpression of two othe r membrane proteins, hZIP1, a human zinc uptake transporter that does not function in yeast (Gaither and Eide, 2001), or ZRT1 2, a nonfunctional mutant of yeast ZRT1 (Gitan et al., 2003), does not effect p ZRT1 -ZRE1 lacZ reporter. This suggests the effect of IZH2 overexpression on pZRT1 -ZRE1 is not caused by aberent protein trafficking (Figure 2-2A). Also, IZH2 overexpression did not repress the p IZH1 -ZRE reporter, which contains a different ZRE (Figure 2-2B). This suggests that IZH2 overexpression did not have a generalized effect on Zap1p activity. We examined the effect of truncated p ZRT1-lacZ reporter constructs on -galactosidase activity. A ZRT1 promoter that contained 361 base pa irs of the promoter was fused to a lacZ construct. This construct contains two known ZREs. ZRE1 was located at to while a

PAGE 52

52 second functional ZRE is located betw een to (Figure 2-2D). The p ZRT1 -361 lacZ construct responds reciprocally to the amount of zinc added to LZM, and IZH2 overexpression repressed of activity of this reporter. A second construct th at contained 305 base pairs of the ZRT1 promoter was used, and was shown to be zinc responsive due to the second functional ZRE; however, overexpression of IZH2 could not repress the pZRT1 -305 lacZ construct (Figure 2-2E). Effect of IZH2 Overexpression on the FET3 Promoter In this paper, we also studied the effect of IZH2 overexpression on a p FET3-lacZ reporter construct. The p FET3-lacZ reporter contains 398 base pairs of the FET3 promoter and responds inversely to the amount of iron added to LIM. The increased activity of -galactosidase, which is due to the transcriptio n factor Aft1p binding to a Fe 3+ R esponse E lement (FeRE) between to (Yamaguchi-Iwai et al., 1996), is repressed by IZH2 overexpression (Figure 2-3A). Ferroxidase activity, the physiolo gical function of Fet3p, also showed full activity during iron deficiency and was repressed by IZH2 overexpression (Figure 2-3B). We also demonstrated that repression of pFET3-lacZ by IZH2 overexpression was incrementally alleviated by decreasing the galactose concentration, which decrea ses expression of genes driven by the GAL1 promoter (Figure 2-3C). Agar plates made iron-defici ent by the addition of fe rrozine were used to compare growth of wild type ce lls carrying an empty vector or IZH2 overexpressor. Wild type cells with the empty vector were capable of growth; however, the cells containing the IZH2 overexpressor did not exhib it growth (Figure 2-3D). We examined the effect of truncated p FET3-lacZ reporter constructs on -galactosidase activity (Figure 2-3E). The p FET3-lacZ reporters containing 297 or 263 base pairs of the FET3 promoter were inducible in ir on-deficient LIM becau se both constructs contained functional FeRE, but were unable to respond to IZH2 overexpression. Also, the p FET3 -FeRE reporter,

PAGE 53

53 which contains only the FeRE from the FET3 promoter inserted into the minimal CYC1 promoter, responds in an iron-depend ent manner, but is not induced by IZH2 overexpression. Both pFET3 -233 and pFET3 -197 constructs were not inducible when grown in iron-deficient LIM (Figure 2-3F). Izh2p was expressed to ensure functional ity; therefore we tagged Izh2p with 3x-HA epitope at the C-terminus The overexpression of IZH2 -HA was able to repress pFET3 -398 similarly to IZH2 overexpression (Figure 2-4A). Furtherm ore, we showed by Western blot that IZH2 -HA construct is galactose i nducible, but not glucose induci ble, when grown in irondeficient LIM (Figure 2-4B). Msn2p and Msn4p Activates FET3 W e were also interested in identif ying activators and repressors of p FET3-lacZ FET3 has been shown to be induced during the diauxic shift in iron replete LIM (Haurie et al., 2003). Since Msn2p and Msn4p are known to activat e gene transcription during this growth phase, we tested their role on FET3 regulation. In msn2 and msn4 mutant strains, ferroxidase activity is constitutively repressed. We also demonstrated that ferroxidase activity was repressed in an msn2msn4 mutant strain compared to the wild type control. To examine Msn2p/Msn4p role during stress response, Mar ilee Weaver used the BY4742 wild type strain containing the pCYC-STRE lacZ reporter and found its ac tivity decreased when it en tered stationary phase. Further examination showed that the p CYC-STRE lacZ reporter was repressed when IZH2 was overexpressed in early log phase (OD600 = 0.04), but not in mid-log phase (OD600 = 1.60). In addition, we examined the effect of Msn2p on the FET3 promoter. To examine the role of Msn2p, we utilized two p FET3-lacZ constructs one containing 398 base pairs and the other 297

PAGE 54

54 base pairs. In iron deficient LIM, results showed p FET3 -298 was activated in the msn2 mutant strain, while pFET3 -398 was constitutively repressed. Nrg1p and Nrg2p Represses FET3 Another scenario in the regulation of FET3 could be the activation of a repressor. Studies fro m Candida albicans have found a CaNrg1p transcriptional repressor may regulate iron uptake genes (Murad et al., 2001). Therefore, we decided to examine the role of S. cerevisiae CaNrg1p homologues, Nrg1p and Nrg2p, in the signaling of FET3 Overexpression of IZH2 in an nrg1 and nrg2 mutant strain resulted in complete loss of repression of p FET3-lacZ activity in irondeficient LIM. Also, the loss of NRG1 and NRG2 genes causes an increase in basal p FET3-lacZ activity. IZH2 overexpression in an nrg1nrg2 mutant strain also led to a loss of repression of ferroxidase activity. In addition, we obtained a p FLO11-lacZ reporter that is known to be negatively regulated by Nrg1p and Nrg2p (van Dyk et al., 2005). IZH2 overexpression was able to repress p FLO11-lacZ activity in iron-deficient LIM. Furthermore, we obtained a GAL1 -driven TAP-tagged Nrg1p and Nrg2p constructs and found that overexpression of Nrg2p-TAP repressed pFET3-lacZ in a wild type strain when grown in iron-deficient LIM. On the other hand, the Nrg1p-TAP overexpressor could no t repress the activity of p FET3-lacZ In an nrg1nrg2 mutant strain, the overexpression of Nrg2p-TAP did not exhibit repression of ferroxidase activity in iron-deficient LIM. The constitutive repr ession of ferroxidase activity in an msn2msn4 mutant strain is alleviated if Nrg1p and Nrg2p both are also kno cked-out. Furthermore, overexpression of IZH2 in the nrg1nrg2msn2msn4 mutant strain exhibited no effect on ferroxidase activity.

PAGE 55

55 Discussion The Izh2p encodes a m embrane protein that belongs to a large and ubiquitous family found in prokaryotes and eukaryot es. This family has been named the PAQR (Progestin and AdipoQ Receptor) family because it contains hu man adiponectin (Yamauchi et al., 2003) and membrane progestin receptors (Zhu et al., 2003a). This family is noted for containing at least seven transmembrane domains and four highly conserved motifs. Also, they have been found to be involved in metal regula tion (Lyons et al., 2004). Repression of FET3 and ZRT1 Involves the C CCTC Regulatory Element In S. cerevisiae the high affinity zinc uptake syst em is controlled by the transcription factor Zap1p by binding to zinc response elements, which induce genes during zinc deficiency such as Zrt1p (Zhao and Eide, 1996). Zrt1p activity was measured by the p ZRT1 -ZRE1 reporter in which a fragment of the ZRT1 promoter containing a known ZRE was inserted into the minimal CYC1 promoter. This reporter construct is inducible by zinc deficiency in a Zap1p inducible manner. However, we showed that overexpression of the Izh2p repressed activity of the promoter construct regardless of nutritional zinc status. We ve rified that the negative effect of Izh2p on lacZ was not an artifact of our experimental design. First, previous studies from our group showing IZH2 overexpression using an estradiol-induc ible system reduces expression of the promoter driven by ZRE1 from ZRT1 during zinc limitation, thus demonstrating the effect was not due to our method of induction. Second, overexpression of two non-functional membrane proteins could not exhibit repression of our lacZ reporter construct, indicating that repression was not due to genera lized effect of protein foldi ng or trafficking by overexpression of a membrane protein (Lyons et al., 2004). Also, previous studi es that exhibited IZH2 overepression did not affect cellular zinc accumulation (Lyons et al., 2000).

PAGE 56

56 Since Zap1p is only known to be regulate d by cellular zinc, we propose that the Izh2p does not target Zap1p. We veri fied this theory by worki ng with another Zap1p-dependent reporter, p IZH1 -ZRE. But, when Izh2p was overexpresse d, repression of this reporter construct was not seen, even though the reporter contai ned a functional ZRE inse rted into the same minimal CYC1 promter, thus confirming IZH2 overexpression does not have a generalized effect on Zap1p activity. Functional ZREs are co ntained in both pZRT1 -ZRE1 and p IZH1 -ZRE and are dependent on Zap1p (Figure 2-1C). But, the exact sequen ce of ZRE and flanki ng bases are slightly different. Thus, we postulated that ZRE1 in the ZRT1 promoter contains an overlapping repressing element that is repressed by overexpression of Izh2p. To test whether IZH2 could regulate the full ZRT1 promoter, we utilized a series of p ZRT1 reporter constructs, one containing 361 nucleotides of the ZRT1 promoter and two ZREs, ZRE1 and ZRE2, while the other reporter construct contai ned 305 nucleotides of the ZRT1 promoter and one ZRE. ZRE2 contains a sequence ACCTTTAAGGT located betw een -203 and -191, while ZRE1 contains a sequence of ACCCTCAAGGT located between -318 to -309. The p ZRT1 -361 and p ZRT1 -305 are both induced during zinc defi ciency; however, induction of p ZRT1 -361 is partially repressed by IZH2 overexpression, but not p ZRT1 -305. This data suggests that the proposed repressive element can be found between -361 and -305 of the ZRT1 promoter and that it most likely overlaps the 29 base pa irs surrounding ZRE1. In investigating the specificity of IZH2 -dependent repression, we discovered that the IZH2 gene also affected the high affinity iron uptak e system. This system is controlled by the transcription factor Aft1p (Rutherford, et al., 2004), which senses cellular iron levels and induces expression of two genes, FET3 and FTR1 whose genes function in tandem to form an oxidase-

PAGE 57

57 permease system that takes iron into the cell during deficiency. In our experiments, pFET3-lacZ reporter construct was used to measure FET3 activity. First, we demonstrated that overexpression of the IZH 2 repressed the p FET3-lacZ reporter. To verify that this was not an artifact of the lacZ construct, we could have performed one of three techniques to answer this question: Northern blot, RT-PCR, or activity of Fet3p. We decided to pursue Fet3p activity because it would show func tionality, and found that IZH2 overexpression decreased Fet3p enzymatic activity in low iron media, confirmi ng an actual loss in the Fet3p phenotype. The levels of repression of p FET3-lacZ and ferroxidase activity were similar to repression caused by iron repletion, suggesting that th e loss in ferroxidase activity is caused by decreased gene expression. We also showed that repression by p FET3-lacZ was not an artifact of our experimental design. First, IZH2 overexpression, which is driven by a GAL1 promoter, exhibited an inability to repress p FET3-lacZ when the medium was incrimentally sw itched from galactose to raffinose, demonstrating repression was due to IZH2 overexpression. We decreased the percentage of galactose, which ramps down the expression of the Izh2p, thereby preventing its effect on FET3 Second, we showed cell viability in iron-limite d LIM is decreased by overexpression of Izh2p, thus confirming that the Izh2p effect is not a physiological irrelevant phenomenon. To identify the repressing element in the FET3 promoter, we obtained a series of p FET3 lacZ reporter constructs, p FET3 -398, pFET3 -297, pFET3 -263, and pFET3-FeRE, which contain the consensus FeRE sequence TGCACCC located between -252 to -245. All p FET3-lacZ constructs responded to cellular iron levels in an Aft1p-depende nt manner (Yamaguchi-Iwai et al., 1996); however, only p FET3 -398 exhibited Izh2p-dependent repression in iron-defecient LIM when IZH2 was overexpressed. Theref ore, we predict the repressing element is between -

PAGE 58

58 398 and -297. When we look closer at the regions surround ing ZRE1 of the ZRT1 promoter and the 100 base pair fragment of the FET3 promoter, which led us to predict the repressing element is a CCCTC motif located at -316 to -312 in the FET3 promoter. This same motif is found overlapping ZRE1 in the ZRT1 promoter at -317 to -313. Nrg1p/Nrg2p and Msn2p/Msn4p Regulate FET3 Two possibilities exis t in which the Izh2p ma y function to repress FET3 One is the inactivation of a co-act ivator. But, at this time, Aft1p is the only known activator of FET3 induction. However, we wanted to know if IZH2 overexpression can affect Aft1p. We tested this theory by using a series of p FET3 lacZ constructs, one containi ng the 398 base pairs and other containing 30 base pairs, loca ted between -263 and -234 of the FET3 promoter, but still containing the Fe3+ Response Element (FeRE). Therefore, the p FET3 -FeRE can still respond to iron deficiency by Aft1p. We demonstrated IZH2 overexpression could repress the p FET3 -398 reporter, but not the pFET3 -FeRE reporter. Thus IZH2 overexpression modulates FET3 activity in an iron-independent manner. Is there there another activator of FET3 ? Haurie, et al showed that FET3 is also induced during the diauxic shift (Haurie et al., 2003). Si nce the stress responsive transcription factors Msn2p and Msn4p are known to activate gene transc ription during the dia uxic shift (Shenhar and Kassir, 2001), we examined th eir role in regulation of FET3 First, we demonstrated that Msn2p and Msn4p are essential for induction of FET3 and that Msn2p/Msn4p may be negatively regulated by IZH2 overexpression. Second, Marilee Weaver showed IZH2 overexpression repressed a stress responsive reporter that is ac tivated by Msn2p and Msn4p. Therefore, we propose Izh2p represses FET3 by inactivating the re pressors, Msn2p and Msn4p. In addition, we were interested in whether the Msn2p/Msn4p effected FET3 regulation by binding to a specific sequence in the FET3 promoter. To test this theory, we used two p FET3-

PAGE 59

59 lacZ constructs in an msn2 strain; one containing 398 base pa irs and the other 297 base pairs. We demonstrated that p FET3 -398 is constitutively repressed in the msn2 but pFET3 -297 responds normally in low iron media. Therefore, FET3 activation by Msn2p mediated by an element located between 398 and 297 base pairs in the FET3 promoter. Nrg1p and Nrg2p Negatively Regulates FET3 A second mechanism of FET3 repression may involve the Iz h2p activating a repressor. This led us to a study performed in Candida albicans showing that strains missing the CaNrg1p transcription factor, exhibited in creased expression of high-affin ity iron uptake genes (Murad et al., 2001). Also, we found that CaNrg1p has two homologues in S. cerevisiae named Nrg1p and Nrg2p. Deletion of either NRG1 or NRG2 results in inability of IZH2 overexpression to repress FET3 Subsequently, we demonstrate that IZH2 overexpression represses FLO11 a gene that is negatively regulated by Nrg1p and Nr g2p. Furthermore, repression of FET3 was also exhibited by Nrg2p overexpression. Repression by Nrg1p/Nr g2p is proposed to occur by binding to a CCCTC consensus sequence in the promoters of targ et genes (Park et al, 1999). This consensus sequence can be found in the FET3 promoter between -316 to -312. Msn2p/Msn4p and Nrg1p/Nrg2p are Epistatic Our data demonstrates that FET3 is positiv ely affected by Msn2p/Msn4p and negatively affected by Nrg1p/Nrg2p. Also, it is proposed that Nrg1p and Nrg2p bind to the consensus sequence CCCTC, which is similar to the CCCCT that serves as a binding site for Msn2p and Msn4p (Park et al, 1999). The similarity between CCCTC and CCCCT consensus sequences make it possible that Nrg1p/Nrg2p and Msn2p/Ms n4p bind to the same consensus sequence. Intriguingly, this idea is bolstered by a recent study suggesting Nrg1p/Nrg2p and Msn2p/Msn4p compete for binding to the same consensus sequen ce in stress-responsive promoters (Vyas et al., 2005). Our data demonstrate that constitutive FET3 repression seen in the msn2 msn4 mutant

PAGE 60

60 strain is alleviated by the additional deletion of NRG1 and NRG2 in the msn2 msn4 nrg1 nrg2 mutant strain. This result suggest s that Nrg1p/Nrg2p is epistatic to Msn2p/Msn4p with respect to FET3 expression. However, at this time, we can not determine if Nrg1p/Nrg2p or Msn2p/Msn4p act on FET3 through cis-regulatory elements. Therefore, we propose Nrg1p/Nrg2p and Msn2p/Msn4p act competitively and antagonistically on FET3 regulation and that Izh2p overe xpression may shift this comp etition by activating Nrg1p/Nrg2p and inactivating Msn2p/Msn4p. Aft1p Activation is not Affected by Mutant Strains In addition to this study, we de monstrated non e of the deletion mutant strains affected FET3 response to the amount of iron added to the me dia, which suggests that the mutant strains do not have a general affect on the ability of iron to inact ivate the Aft1p iron-responsive transcriptional activator. Therefore, this response allows us to measure IZH2 overexpression in these mutant strains. Summation Our findings support two conclusions. First, Izh2p activates a signal transduction pathway that represses both ZRT1 and FET3 expression. Second, Izh2p affects iron hom eostasis via the Nrg1p/Nrg2p and Msn2p/Msn4p. Future experiments will focus on investigating the physical binding of NRG to the promoter s and putative binding sites in question.

PAGE 61

61 inactive (cyto)Nrg1p/Nrg2p Nrg1p/Nrg2p active inactive AMPK AMPKactiveMsn2p/Msn4p Msn2p/Msn4pactive active active inactivePKA PKA active inactivePkh1p/Pkh2pPkh1p/Pkh2p ?Phytoceramide Phytosphingosine Ligand (cyto) (nucl) (nucl) (nucl) (cyto) (cyto) (nucl) CCCTC FET3 Figure 2-1. A model for the mechanism of FET3 by IZH2 overexpression. Solid arrows indicate protein activation or localization of pr otein. Broken arrows indicate protein interaction on localization a nd activation. Cyto = protein is cytoplamic. Nucl = protein is nuclear. Shaded blue portions represent areas studied in chapter.

PAGE 62

62 -Galactosidase Activity % of WT Low Zn2+ Vector Control 0 20 40 60 80 100 120 140 Control hZIP ZRT1 2 Izh2p A M [Zn2+] 1.0 1050 500 100 1,000 Figure 2-2. Repression of ZRT1 by IZH2 -overexpression. BY4742 wild t ype strain is used in panels A and B, while DY1457 wild type strain was used in panel E. Galactosidase activities are shown as a per centage of fully induced activity in wild type strain carrying an empt y control vector grown in iron-deficient LIM. The error bars represent 1 standard deviation for experiments performed in triplicate. (A) Galactosidase activity of y east strain carrying the p ZRT1 -ZRE2 and the IZH2 overexpressors individually. Cells are gr own in LZM/galactose supplemented with 1 to 1,000 M Zn2+. (B) -Galactosidase activity of yeast strain carrying the p IZH1 ZRE and the IZH overexpressors individually. Ce lls are grown in LZM/galactose supplemented with 1 and 1,000 M Zn2+. (C) Sequences of p ZRT1 -ZRE1 and p IZH1 ZRE inserted into the minimal CYC1 promoter. CCCTC is the putative binding site overlapping p ZRT-ZRE1. (D) Picture of ZRT1 plasmid constructs: p ZRT1 -361 and pZRT1 -305. Numbers represent nucleotides in ZRT1 promoter. (E) -Galactosidase activity of yeast strain carrying the p ZRT1 -361 or pZRT1 -305 and the IZH2 overexpressor. Cells are grown in LZM /galactose supplemented with 1 and 1,000 M Zn2+.

PAGE 63

63 Galactosidase Activity % of Low Zn2+ Vector Control 0 20 40 60 80 100 120 140 p ZRT1 -ZRE1 p IZH1ZRE Wild Type Izh2pB C CCAAAGATA CCCTCAAGGTTCTCATCTGT ATCATGCAACCTTTA GGGTCCAAGCCCTTp IZH1 -ZRE p ZRT1 -ZRE1 D Figure 2-2 continued. lacZZRE1 ZRE2-361 lacZZRE2 -305ZRT1 ZRT1

PAGE 64

64 Galactosidase Activity % of Low Zn2+ Vector Control 0 20 40 60 80 100 120 p ZRT1 -361 p ZRT1 -305 1.0 1,000 1.0 1,000+ +IZH2 OX Zn2+ ( M)E Figure 2-2 continued.

PAGE 65

65 -Galactosidase Activity % of WT Low Fe3+ Vector Control 0 20 40 60 80 100 120 140 Control Izh2p [Fe3+] 1.0 1050 100 500 1,000A Figure 2-3. Repression of FET3 by IZH2 -overexpression. BY4742 wild t ype strain is used in all panels. -Galactosidase (Panels A, C, F, and G) or ferroxidase (Panel B) activities are shown as a percentage of fully induced activity in wild type strain carrying an empty control vector grown in iron-defici ent LIM. The error bars represent 1 standard deviation for experiment s performed in triplicate. (A) -Galactosidase activity of yeast strain carrying the p FET3 -398 and the IZH overexpressors individually. Cells are grown in LIM/ galactose supplemented with 1 to 1,000 M Fe3+. -Galactosidase activity is induced at low concentrations of iron, but not at high concentrations of iron in the BY4742 wild type strain; however, IZH2 overxpression represses p FET3 -398 -galactosidase activity. (B) Cell surface ferroxidase activity is induced by low iron in BY4742 wild type strain carrying an empty vector, but not in yeast carrying the IZH overexpressors (C) Izh2p overexpression on p FET3 -398 decreases as the % galactose in iron-d eficient LIM is decreased. (D) Izh2p overexpression causes a growth defect in pl ates containing synthe tic galactose media supplemented with 1 mM ferrozine and 20 M Fe3+. (E)Pictures of p FET3 -lacZ constructs: p FET3 -398, pFET3-297, p FET3 -263, pFET3 -233, pFET3 -197 and pCYC1 -FeRE. Numbers represent nucleotides in FET3 promoter. (F) Galactosidase activity is indu ced in strains carrying p FET3 -398, pFET3 -297, pFET3 263, or pCYC1 -FeRE in iron-deficient LIM. Izh 2p overexpression in iron-deficient LIM represses only the p FET3 -398. (G) -Galactosidase activity is induced in strains carrying p FET3 -398, but not with p FET3 -233 or pFET3 -197 in iron-deficient LIM. Izh2p overexpression in iron-defici ent LIM represses only the p FET3 -398.

PAGE 66

66 Ferroxidase Activity % of WT Low Fe3+ Vector Control 0 20 40 60 80 100 120 1 M Fe3+ 1,000 M Fe3+ Control Izh2pB 0.00 -Galactosidase Activity % of WT Low Fe3+ Vector Control 0 20 40 60 80 100 120 Izh2p 0.01 0.05 0.10 0.30 0.50 0.75 1.00 1.50 2.00 % GalactoseC Figure 2-3 continued.

PAGE 67

67 D E CYC Figure 2-3 continued. Izh2p Izh2p with Ferrizine

PAGE 68

68 -Galactosidase Activity % of WT Low Fe3+ Vector Control 0 20 40 60 80 100 120 p FET3 -398 p FET3 -297 p FET3 -263 p CYC -FeRE 1.0 1,000 1.0 1,000+ +IZH2 OX Fe3+ ( M)F -Galactosidase Activity % of WT Low Fe3+ Vector Control 0 20 40 60 80 100 120 p FET3 -398 p FET3 -233 p FET3 -197 1.0 1,000 1.0 1,000+ +IZH2 OX Fe3+ ( M)G Figure 2-3 continued.

PAGE 69

69 Control -Galactosidase Activity % of WT Low Fe3+ Vector Control 0 20 40 60 80 100 120 1 M Fe3+ 1 mM Fe3+ Izh2pIzh2p-3HAA Figure 2-4. Izh2p-3HA expression. BY4742 wild type strain is used in panels A to C. Galactosidase activities are shown as a per centage of fully induced activity in wild type strain carrying an empt y control vector grown in iron-deficient LIM. The error bars represent 1 standard deviation for experiments performed in triplicate. (A) Galactosidase activity of p FET3 -398 is repressed in a strain carrying Izh2p or Izh2p3HA overexpressors in iron-deficient LI M. (B) Western blot Izh2p-3HA tagged protein is induced in a strain grown in galact ose but not in a strain grown in glucose. (C) Western blot showing Izh2p-3HA tagged protein is expressed in the strain carrying the Izh2p-3HA plasmid but not in the strain carrying the Izh2p plasmid.

PAGE 70

70 Ferroxidase Activity % of WT Low Fe3+ Vector Control 0 20 40 60 80 100 120 Wild Type msn2msn4msn4msn2A Figure 2-5. MSN2 and MSN4 positively regulate FET3 BY4742 wild type strain is used in panels A to C, while the wild type strain MCY5326 and msn2msn4 mutant strain were used in panel B. msn2 and msn4 mutant strains were used in panel A. Galactosidase (Panels B and C) or ferroxida se (Panel A) activities are shown as a percentage of fully induced activity in w ild type strain carry ing an empty control vector grown in iron-deficient LIM. The error bars represent 1 standard deviation for experiments performed in triplicate. (A ) Cell surface ferroxidase is repressed in msn2, msn4, and msn2msn4 mutant strains. (B) p CYC-STRE lacZ activity is repressed when IZH2 is overexpressed at OD600 0.04, but not at OD600 1.60. (C) pFET3 -398 activity was constitutively repressed in the msn2 mutant strain, while pFET3 -297 activity was induced.

PAGE 71

71 Wild Type -Galactosidase Activity % of WT Low Fe3+ Vector Control 0 20 40 60 80 100 120 1 M Fe3+ 1 mM Fe3+ Wild Type msn2msn2 p FET3 -398 p FET3-297C Figure 2-5 continued

PAGE 72

72 -Galactosidase Activity % of WT Low Fe3+ Vector Control 0 20 40 60 80 100 120 140 160 180 Control Izh2p Wild Type nrg1nrg2A Figure 2-6. Nrg1p and Nrg2p are essential for repr ession. BY4742 wild type strain was used in panels A and D. MCY5326 wild type strain was used in panels B, E, and F. 1278b wild type is used in panel C. nrg1 (Panel A), nrg2 (Panel A), nrg1 nrg2 (Panels B, E, and F), msn2 msn4 (Panel F), and nrg1 nrg2 msn2 msn4 (Panel F) strains were also used. -Galactosidase (Panels A, C a nd D) or ferroxidase (Panels B, E, and F) activities are shown as a percen tage of fully induced activity in wild type strain carrying an empty cont rol vector grown in iron-defi cient LIM. The error bars represent 1 standard deviation for expe riments performed in triplicate. (A) IZH2 overexpression does not affect p FET3-lacZ activity in an nrg1 and nrg2 mutant strains. (B) Ferroxidase activity is not repressed by IZH2 in an nrg1 nrg2 mutant strain. (C) IZH2 overexpression represses a FLO11-lacZ construct. (D) The TAPtagged Nrg2p overexpressor, but not TA P-tagged Nrg1p overexpressor, repressed pFET3-lacZ (E) Ferroxidase activity is not affected by TAP-tagged Nrg2p overexpression in an nrg1 nrg2 mutant strains (F) Ferroxidase was constititively repressed in msn2 msn4 mutant strain, partia lly repressed in an nrg1 nrg2 msn2 msn 4 mutant strain, and repre ssion is alleviated in an nrg1 nrg2 mutant strain. IZH2 overexpression had no affect in these strains.

PAGE 73

73 Ferroxidase Activity % of WT Low Fe3+ Vector Control 0 20 40 60 80 100 120 140 Control Izh2p nrg1nrg2Wild TypeB Control -Galactosidase Activity % of Low Fe3+ Vector Control 0 20 40 60 80 100 120 Izh2pC Figure 2-6 continued.

PAGE 74

74 -Galactosidase Activity % of WT Low Fe3+ Vector Control 0 20 40 60 80 100 120 Wild TypeTAP-Tagged Nrg1p TAP-Tagged Nrg2pD Wild Type Ferroxidase Activity % of WT Low Fe3+ Vector Control 0 20 40 60 80 100 120 140 Control TAP-Tagged Nrg2p nrg1nrg2E Figure 2-6 continued.

PAGE 75

75 Wild Type Ferroxidase Activity % of WT Low Fe3+ Vector Control 0 20 40 60 80 100 120 140 Control Izh2p msn2msn4nrg1nrg2nrg1nrg2msn2msn4F Figure 2-6 continued.

PAGE 76

76 -Galactosidase Activity % of WT Low Fe3+ Vector Control 0 20 40 60 80 100 120 140 1 M Fe3+ 1 mM Fe3+ ABY4742 Wild Type nrg1nrg2 Figure 2-7. Aft1p responds normally to iron-defeci ency in most mutant strains. BY4742 wild type strain was used in panels A and B. MCY5326 wild type strain was used in panel C. nrg1 (Panel A), nrg2 (Panel A), msn2 (Panel B), msn4 (Panel B), nrg1 nrg2 (Panel C), msn2 msn4 (Panel C), and nrg1 nrg2 msn2 msn4 (Panel C) mutant strains were also used. -Galactosidase (Panel A) or ferroxidase (Panels B and C) activities are shown as a pe rcentage of fully induced activity in wild type strain carrying an empt y control vector grown in iron-deficient LIM. The error bars represent 1 standard deviation for e xperiments performed in triplicate. (A) Galactosidase activity is indu ced in iron-defeciency in the BY4742 wild type strain, and in strains lacking NRG1 and NRG2 (B) Ferroxidase activity is not induced in strains lacking msn2 and msn4 (C) Ferroxidase activ ity is induced in irondefeciency in the MCY5326 wild type strain, or in strains lacking NRG1NRG2, and MSN2MSN4NRG1NRG2 Ferroxidase activity is not induced in the strain lacking MSN2MSN4

PAGE 77

77 BY4742 Wild Type Ferroxidase Activity % of WT Low Fe3+ Vector Control 0 20 40 60 80 100 120 1 M Fe3+ 1 mM Fe3+ Bmsn2msn4 MCY5326 Wild Type Ferroxidase Activity % of WT Low Fe3+ Vector Control 0 20 40 60 80 100 120 140 1 M Fe3+ 1 mM Fe3+ msn2msn4nrg1nrg2msn2msn4nrg1nrg2C Figure 2-7 continued

PAGE 78

78 CHAPTER 3 THE SIGNALING OF FET3 REPRESSION BY IZH2 OVEREXP RESSION Introduction Chapter 2 details experiments showing that IZH2 overexpression represses FET3 through the carbon source-dependent transcription repres sors, Nrg1p and Nrg2p (Berkey et al., 2004) and the stress-responsive transcripti onal activators, Msn2p and Msn4p (Smi th et al., 1998). We also demonstrated that Nrg1p/Nrg2p and Msn2p/Msn4p likely compete for the same binding site on the FET3 promoter that is essential for IZH2 -dependent repression of FET3. In this chapter, we identified signaling proteins upstream of these transcription factors, which are required for FET3 repression by IZH2 overexpression. We show that IZH2 -dependent repression of FET3 requires protein kinase A (PKA), which inhibits Msn2p/Msn4p (Smith et al., 1998), and AMP-dependent kinase (AMPK), which inhibits Nrg1p/Nrg2p (Kuchi n et al., 2002). Finally, we demonstrate that IZH2 -dependent repression requires the sphingoid base sensing kinases Pkh1p and Pkh2p (Figure 3-1). Thus, these results suggest that IZH2 overexpression activates PKA and inactivates AMPK that leads to repression of FET3 Materials and Methods Yeast Strains and Plasmids The wild type (BY4742) strain and the pkh1 pkh2 tpk1 tpk2 tpk3 ras2 tos3 elm1 sak1 snf1 snf4 gal83 sip1 and sip2 mutant strains were obtained from Euroscarf and their genotypes are listed in Supplementary Table 1. The p FET3-lacZ plasmid (Gift of D. Eide) has the promoter region of the p FET3 gene fused to the lacZ open reading frame and is URA3 selectable. IZH2 was cloned from yeast genomic DNA into pRS316GAL1 ( LEU2 ) via homologous recombination as previously descri bed (Lyons et al., 2004). The primers used to

PAGE 79

79 clone IZH2 are listed in Supplementary Table 2. All clones were sequenced to ensure that no errors occurred during PCR. -Galactosidase and Ferroxidase A ssays -Galactosidase and ferroxidase assays were performed as described in chapter 2 except for the addition of the following compounds to LIM: adenosine 3, 5-cyclic monophosphate sodium salt monohydrate (cAMP) (Sigma #A6885) or 5-aminoimidazole-4-carboxamide 1-Dribofuranoside (AICAR) (Sigma #A9978) were added to the cultures at various concentrations. Results Fet3p is a cell surface ferroxidase that regu lates high-affinity ir on uptake (Singh et al., 2006). Overexpression of IZH2 causes repressio n of FET3 activity; however, the signaling mechanism of how this occurs remains a mystery. We have alrea dy demonstrated that IZH2 dependent repression of FET3 occurs through a competition between Nrg1p/Nrg2p and Msn2p/Msn4p. Thus, we examined the involve ment of PKA and AMPK complexes, the upstream signaling proteins of bot h Nrg1p/Nrg2p and Msn2p/Msn4p, respectively. PKA and Ras-cAMP is involved in FET3 regulation Previous studies have shown that PKA negatively regulates Msn2p/Msn4p activity and represses Aft1p-target genes (Robe rtson et al., 2000). PKA is m ade of three catalytic isoforms (Tpk1p, Tpk2p, and Tpk3p) and deletion of any of thes e isoforms results in loss of repression of pFET3-lacZ in iron-deficient LIM caused by IZH2 (Figure 3-2A). The PKA complex is known to be positively regulated by th e Ras2p-cAMP pathway (Jiang et al., 1998). This pathway involves Ras2p stimulating adenylate cyclase to produce cAMP, which, in turn activates PKA. Therefore we examined a strain missing RAS2, and found pFET3-lacZ repression by IZH2 overexpression was also alleviated (Figure 3-2A). When cAMP was added to the ras2 mutant strain, IZH2 overexpression was able to restore repression of p FET3-lacZ (Figure 3-2B).

PAGE 80

80 AMPK is involved in FET3 Regulation AMP-dependent kinase (AMPK) is a negativ e regulator of Nrg1p and Nrg2p (Kuchin et al., 2002) and has been implicated in the regulation of iron-regulated genes (Haurie 2003 JBC). The AMPK com plex is heterotrimer, which contains a catalytic -subunit (Snf1p), a stimulatory -subunit (Snf4p), and one of three -subunits (Gal83p, Sip1p, and Si p2p) that regulate AMPK cellular localization (Vin cent et al., 2001). AMPK is activat ed by upstream kinases, AMPKK, of which there are three isofor ms (Elm1p, Tos3p, and Sak1p) (Hedbacker et al., 2004). Thus, we decided to examine if th e AMPK subunits were involved in IZH2 -dependent repression of pFET3-lacZ We show that strains lacking Snf1p results in constitutive repression of ferroxidase activity (Figure 3-3A), while th e loss of Gal83p or Sak1p, which are responsible for nuclear AMPK activity, also exhi bits constitutive repression of p FET3-lacZ (Figure 3-3B). In addition, strains lacking the and cytoplasmic subunits of AMPK (Snf4p, Sip1p, and Sip2p) resulted in a slight decrease in p FET3 lacZ Moreover, in the snf4 sip1 and sip2 mutant strains, overexpression of IZH2 was unable to repress p FET3-lacZ activity (Figure 3-3C). Furthermore, loss of Tos3p and Elm1p, two of the AMPKK isoforms, exhibited constitutive repression of pFET3-lacZ (Figure 3-3D). In this study, we also examined a known activator of mammalian AMPK, AICAR (Corton et al., 1995). When this molecu le was added to our media, it produced a decrease in pFET3-lacZ activity (Figure 3-3E), suggesting that AMPK may play a role in signal transduction. FET3 Regulation Requires PKH1 and PKH2 PKA has been shown to negatively regulate Msn2p/Msn4p. What, the n, regulates P KA? One possibility is Pkh1p and Pkh2p. Pkh1p and P kh2p are a pair of partially redundant kinases that are the yeast homologues of mammalia n phosphoinositide-dependent kinase (PDK) (Casamayor et al, 1999). Deletion of Pkh1p a nd Pkh2p resulted in nearly complete loss of IZH2 -

PAGE 81

81 dependent repression of p FET3-lacZ although Pkh1p seemed to have a greater role than Pkh2p (Figure 3-4A). Mutant Strains do not Effect Aft1p Regulation We were also interested to s ee if mutant strains affected the response of Aft1p to iron. Aft1p is a transcription factor that senses the availability of iron in the media (Dancis et al., 1992). Therefore, Aft1p should be turned on dur ing iron deficiency and turned off during iron repletion. Our results dem onstrate that deletion of PKH1 PKH2 RAS2, TPK1 TPK2 or TPK3 exhibited induction of p FET3-lacZ (Figure 3-5A) in iron-defecien t LIM. Also, yeast missing SNF4 SIP1 or SIP2 (Figure 3-5B) demonstrated partial induction of p FET3-lacZ whereas the loss of GAL83 SAK1, TOS3, ELM1 (Figure 3-5B) or SNF1 (Figure 3-5C) showed repression of pFET3-lacZ or ferroxidase activity in iron-defecient LIM. Discussion In this study, we showed that Izh2p affects the regulation of FET3 More im portantly, we provide evidence that suppor ts a new mechanism for FET3 regulation. We intitially demonstrated that IZH2 overexpression represses FET3 by activating Nrg1p/Nrg2p and inactivating Msn2p/Msn4p Thus, we proceeded to identify the upstream signaling proteins involved in regulation of NRG and MSN. AMPK One way in which IZH2 could affect the compe tition between Nrg1p/Nrg2p and Msn2p/Msn4p is by activating Nrg1p/Nrg2p, as shown in chapter 2. The zinc finger transcriptional repressor proteins Nrg1p and Nrg2p are known to inte ract physically with and are negatively regulated by AMPK (Vyas et al., 2001). Therefore, it is possible that Izh2p activates Nrg1p/Nrg2p by inactivating AMPK.

PAGE 82

82 The role of AMPK is complicated by its multiple functional isoforms that depend upon the cellular localization and regul atory subunits. Moreover, AMPK is activated by one of three AMPK kinases (Sak1p, Elm1p, and Tos3p) (Elbing et al, 2006). In yeas t, most of AMPKs known functions occur in the nucleus, including Nrg1p/Nrg2p inactivation. We present evidence that nuclear AMPK is required for FET3 transcription in low iron. This model is supported by the fact that strains lacking SNF1 exhibited constitutive low FET3 expression even in medium containing low iron. Furthermore, strains lacking GAL83 and SAK1 which are responsible for active nuclear AMPK, exhibit the same constitutive FET3 repression. We designed experiments to test if cytoplasmic AMPK has a role in IZH2-dependent repression. Evidence for a role of cyt oplasmic AMPK comes from data showing FET3 repression by the addition of an AMPK activ ator (AICAR). Also, strains lacking the and cytoplasmic -subunits of AMPK (Snf4p, Sip1p, a nd Sip2p) are insensitive to IZH2 -dependent repression. In addition, strains lacking any one of the three AMPK activating kinases exhibit constitutive FET3 repression. These data support a model in which IZH2 overexpression has opposite effects on nuclear and cyto plasmic isoforms of AMPK. PKAs Role in IZH2 -Dependent Repression. Studies in yeast have shown that PKA re presses Msn2p/Msn4p and regulates the growth and stress response (Smith et al., 1998). Furt hermore, Izh2p has also been implicated in signaling through the Ras-cAMP/PKA pathway (Narasimhan et al., 2005). In addition, PKA is reported to regulate genes involved in iron homeostasis (Rober tson et al., 2000). Finally, a report has also demonstrated that active P KA causes Snf1p to build-up in the cytoplasm (Hedbacker, et al., 2004) where it would have li mited access to Nrg1p/Nrg2. This hypothesis is supported by our data which demonstrates the lo ss of any of the PKA catalytic subunits (Tpk1p, Tpk2p, and Tpk3p) eliminates IZH2 -dependent repression of FET3

PAGE 83

83 We have also shown FET3 repression via Izh2p requi res Ras2p-cAMP signaling. However, is the Ras2p or production of cAMP requi red? We have demonstrated that the Ras2p is not essential, but the a ddition of cAMP was able to re-establish the ability of IZH2 to repress FET3 in a ras2 mutant strain. This data suggests that Izh2p is not directly coupled to Ras2p and is only required for the act ivation of PKA via cAMP. Pkh1p/Pkh2p as a Master Regulator If IZH2 -dependent repression of FET3 involves PKA, then what are the upstream proteins that transduce the signal? We suspected the involvement of a pair of partially redundant kinases known as Pkh1p and Pkh2p. This idea came from the fact that yeast PAQR homologues, IZH1 and IZH4 are divergently transcribed from PKH1 and PKH2 Divergently transcribed genes often belong in the same chemical pa thway (Kruglyak and Tang, 2000). Also, Pkh1p and Pkh2p are homologues of the mammalian phosphoinositide-dependent pr otein kinase (PDK) (Casamayor et al, 1999). PDK is a central regu lator that activates downstream kinases in the AGC superfamily (Vanhaesebroeck and Alessi 2000). Thus, we looked at strains missing PKH1 or PKH2 and found that FET3 expression could not be repressed by IZH2 overexpression. Even though Izh2p elicit a minimal effect in the pkh1 and pkh2 single mutant, we were unable to investigate the additive effect of losing both genes because the pkh1 pkh2 double mutant strain is lethal (Inagaki et al.,1999). Furthermore, two-hybrid scr eens have demonstrated physical interactions between Pkh1p and the P KA isoform, Tpk3p (Ho et al., 2002). Summation Based on our data, we propose a model for signal transduction via IZH2 overexpression that results in repression of hi gh affinity iron uptake genes in yeast. Our data show that IZH overexpression activates Pkh1p/P kh2p. Pkh1p/Pkh2p then regulates the activity of PKA and

PAGE 84

84 AMPK and ultimately activates the transcri ption factors Nrg1p/Nrg2p and inactivates Msn2p/Msn4p.

PAGE 85

85 inactive (cyto)Nrg1p/Nrg2p Nrg1p/Nrg2p active inactive AMPK AMPKactiveMsn2p/Msn4p Msn2p/Msn4pactive active active inactivePKA PKA active inactivePkh1p/Pkh2pPkh1p/Pkh2p ?Phytoceramide Phytosphingosine Ligand (cyto) (nucl) (nucl) (nucl) (cyto) (cyto) (nucl) CCCTC FET3 Figure 3-1. A model for the mechanism of FET3 by IZH2 overexpression. Solid arrows indicate protein activation or localization of pr otein. Broken arrows indicate protein interaction on localization a nd activation. Cyto = protein is cytoplamic. Nucl = protein is nuclear. Shaded blue portions represent areas studied in chapter.

PAGE 86

86 -Galactosidase Activity % of WT Low Fe3+ Vector Control 0 20 40 60 80 100 120 140 Control Izh2p Wild Type ras2tpk1tpk2tpk3A Figure 3-2. PKA and Ras-cAMP is involved in FET3 regulation. BY4742 wild type and ras2 mutant strains are used in panels A and B. tpk1, tpk2, and tpk3 mutant strains are used in panel A. -Galactosidase activities are sh own as a percentage of fully induced activity in wild t ype strain carrying an empty control vector grown in irondeficient LIM. The error bars represen t 1 standard deviation for experiments performed in triplicate. -Galactosidase activity of yeast strain carrying the p FET3 lacZ and the IZH overexpressors individually. (A) IZH2 overexpression did not repress -Galactosidase activity in tpk1, tpk2, tpk3, and ras2 mutant strains. (B) IZH2 overexpression in a ras2 mutant strain treated wi th 3 mM cAMP repressed pFET3-lacZ activity.

PAGE 87

87 Hours of exposure to 3 mM cAMP prior to assay 0 5 10 15 20 25 -Galactosidase Activity % of WT Low Fe3+ Vector Control 40 50 60 70 80 90 100 110 120 WT Control ras2 Control ras2 Izh2p B Figure 3-2 continued.

PAGE 88

88 Ferroxidase Activity % of WT Low Fe3+ Vector Control 0 20 40 60 80 100 120 Wild Type snf1A Figure 3-3. AMPK is involved in FET3 regulation. BY4742 wild type strain was used in all panels. snf1 snf4 gal83 sip1 sip2 sak1 tos3 and elm1 mutant strains were also used. -Galactosidase (Panels B to E) or ferroxidase (Panel A) activities are shown as a percentage of fully induced activity in wild type strain carrying an empty control vector grown in iron-defici ent LIM. The error bars represent 1 standard deviation for experiments performed in triplicate. (A) Ferroxidase activity is constitutively repressed in snf1 strain. (B) Deletion of GAL83 and SAK1 genes shows constitutive repression of pFET3-lacZ (C) Deletion of SNF4 SIP1 and SIP2 genes exhibit partial repression of p FET3-lacZ IZH2 overexpression had no affect in these strains. (D) Deletion of TOS3 and ELM1 genes showed constitutive repression of pFET3-lacZ (E) Addition of 500 M AICAR results in partial repression of pFET3-lacZ

PAGE 89

89 -Galactosidase Activity % of WT Low Fe3+ Vector Control 0 20 40 60 80 100 120 Control Izh2p Wild Typegal83sak1B -Galactosidase Activity % of WT Low Fe3+ Vector Control 0 20 40 60 80 100 120 Control Izh2p Wild Type snf4sip1sip2C Figure 3-3 continued.

PAGE 90

90 -Galactosidase Activity % of WT Low Fe3+ Vector Control 0 20 40 60 80 100 120 Wild Type tos3elm1D Untreated -Galactosidase Activity % of WT Low Fe3+ Vector Control 0 20 40 60 80 100 120 AICARE Figure 3-3 continued.

PAGE 91

91 -Galactosidase Activity % of WT Low Fe3+ Vector Control 0 20 40 60 80 100 120 140 Control Izh2p Wild Type pkh1pkh2A Figure 3-4. FET3 regulation requires PKH1 and PKH2 BY4742 wild type strain was used in panel A. pkh1 and pkh2 mutant strains are used in panel A. -Galactosidase activities are shown as a percentage of fully induced activity in wild type strain carrying an empty control vector grown in iron-deficient LIM. The error bars represent 1 standard deviation for e xperiments performed in triplicate. Galactosidase activity of y east strain carrying the p FET3-lacZ and the IZH overexpressors individua lly. (A) Deletion of PKH1 and PKH2 genes results an inability of the IZH2 to repress pFET3-lacZ

PAGE 92

92 -Galactosidase Activity % of WT Low Fe3+ Vector Control 0 20 40 60 80 100 120 140 1 M Fe3+ 1 mM Fe3+ AWild Type pkh1pkh2ras2tpk1tpk2tpk3 Figure 3-5. Aft1p responds normally to iron-defeci ency in most mutant strains. BY4742 wild type strain was used in panels A, B, and C. -Galactosidase (Panel A and B) or ferroxidase (Panel C) activiti es are shown as a percentage of fully induced activity in wild type strain carrying an empty control vector grown in iron-deficient LIM. The error bars represent 1 standard deviation for experiments performed in triplicate. (A) -Galactosidase activities are induced in the BY4742 wild type stain, or strains lacking PKH1 PKH2 RAS2, TPK1 TPK2 or TPK3 (B) -Galactosidase activities are partial induced in strains lacking SNF4 SIP1 or SIP2 but not in st rains lacking GAL83 SAK1, TOS3, or ELM1 (C) Ferroxidase activity is induced in the BY4742 wild type stain, but not in strains lacking SNF1

PAGE 93

93 -Galactosidase Activity % of WT Low Fe3+ Vector Control 0 20 40 60 80 100 120 1 M Fe3+ 1 mM Fe3+ Wild Type sip1sip2gal83sak1tos3elm1snf4B Wild Type Ferroxidase Activity % of WT Low Fe3+ Vector Control 0 20 40 60 80 100 120 1 M Fe3+ 1 mM Fe3+ Csnf1 Figure 3-5 continued.

PAGE 94

94 CHAPTER 4 PAQR-DEPENDENT REPRESSION OF FET3 IS MEDIATED BY SPHINGOLIPIDS Introduction Sphingolipids, along with sterols, are im portant components of cell membranes in eukaryotic cells (Lisman et al., 2004). Th eir metabolites ceramide, sphingosine, and sphingosine-1-phosphate have important roles in cell biology (Merrill et al., 1989), particularly, their role in intracellular signaling. Sphingolipids, structurally, are amphipathic molecules that have hydrophobic and hydrophilic components. The hydrophobic region consists of a sphingoid long chain base (sphingosine in mammals or phytosphingosine in plants and fungi) to which a fatty acid is attached by an amide linkage to form a ceramide, which is the simplest sphingolipid. More complex sphingolipids have a polar head group consisting of a carbohydrate that is attached by glycosidic linka ge to the ceramide at the C1-OH group are known as sphingoglycolipids. Ceramides and sphingosine exert opposite effects in biological systems. Ceramide play a role in cellular growth, growth arrest, and stress responses (Merri ll et al., 1997; Hannun and Obeid, 1997; Hannun, 1997; Hannun, 1996), while sphingosin e is involved in stimulating growth and suppressing apoptosis (Cuvillier, 2002). No t surprisingly, the ra tio of ceramide to sphingosine must be tightly regulated inside the cell. This ratio can be altered by enzymes. One such group of enzymes, ceramidases, plays an intricate role in mainta ining a balance of this ratio. In Saccharomyces cerevisiae, the alkaline ceramidase, Ypc 1p, deacylates phytoceramide to phytosphingosine and a free fatty acid. In additio n, Ypc1p exhibits reverse activity forming phytoceramide from the free fatty ac id and phytosphingosine (Hannun, 1996).

PAGE 95

95 Ypc1p is structurally related to the PAQRs (Figure 4-1) (Lyons et al., 2004). Both the ceramides and PAQRs contain at least seven tran smembrane domains (Figure 4-2) and a similar predicted topology with the cytopl asmic N-terminus and extra-cytoplasmic C-terminus. This predicted toplogy has been confirmed for Adi poR1, AdipoR2, Izh2p and Izh4p (Yamauchi et al., 2003; Kim et al., 2003; Kim et al, 2006). Since Ypc1p and the PAQR proteins are related, we investigated whether th e Izh proteins also possess ceramidase activity. In this study, we show that increasing endogenous sphingoid bases by overexpressing Ypc1p results in FET3 repression by a mechanism that is similar to the IZH s. Furthermore, we show that adding exogenous sphingoi d bases have the same effect on FET3 Finally we showed that the ceramidase inhibitor, D-erythro MAPP (Bielawska et al., 1996), alleviated FET3 repression caused by IZH overexpression. Thus, we propose that PAQR overexpression generates a sphingolipid second messenger (Figure 4-3) responsible fo r the downstream effects on FET3 Materials and Methods Yeast Strains and Plasmids The W 303-1A wild type strain and ypc1 ydc1 strain were a gift of Dr. Howard Rielzman (University Basel, Switz erland). Wild type (BY4742) strain and pkh1 pkh2 tpk1 tpk2 tpk3 ras2 msn2 msn4 snf1 snf4 gal83 sip1 sip2 nrg1, nrg2, lcb4 lcb5 sur1 sur2 sur4 fen1 scs7 csg2 and ipt1 mutant strains were obtained from Euroscarf and their genotypes are liste d in Supplementary Table 1. The p FET3-lacZ plasmids (Gift of D. Eide) has th e promoter region of the FET3 gene fused to the lacZ open reading frame and is URA3 or LEU2 selectable. IZH2 was cloned from genomic DNA into pRS316GAL1 ( LEU2 or URA3), while YPC1 was cloned fr om genomic DNAinto pRS316-GAL1 ( URA3 ) via hom ologous recombination as previously describe d (Lyons et al., 2004). Primers used to clone

PAGE 96

96 IZH2 and YPC1 are listed in Supplementary Table 2. All clones were sequenced to ensure that no errors occurred in PCR. -Galactosidase and Ferroxidase A ssays -Galactosidase and ferroxidase sssays were performed as described in chapter 2 except for the addition of the following compounds to LIM. Fr esh stocks of sphingosin e (SPH) (Avanti Polar Lipids #860490), phytosphingosine (PHS) (Avan ti Polar Lipids #860499), dihydrosphingosine (DHS) (Avanti Polar Lipids #860498), and stearylamine (Sigma-Aldrich #124-30-1) were dissolved 1:1 in chloroform/methanol, fumonisin B1 (FB1) (Cayman #116355-83-0) in methanol, sphingosine kinase inhibitor 2 (SPH KI) (C ayman #312636-16-1) in dimethylsulfoxide, and N,N-dimethylsphingosine (N,N-DMS) (Cay man #119567-63-4), D-erythro-MAPP (Cayman #143492-38-0), and myriocin (Cayman #35891-70-4) in ethanol. For all experiments equal volumes of solvent were added to untreated cells to control for vehicle effects. For all cases, except myriocin, the drug was added upon reinoculat ion in LIM. In the case of myriocin, cells were treated with the drug for one hour prior to -galactosidase or ferroxidase assay. Results Exogenous and Endogenous Sphingoid Bases Repress FET3 Since sphingoid bases are the only known ac tivator of yeast Pkh1p and Pkh2p (Liu, et al., 2005), we decided to test their effect on FET3 regulation. In this study, we show that the addition of 100 M sphingosine, phytosphingosine, and dihydrosphingosine resulted in repression of pFET3-lacZ However, another way to increase the concentration of sphingoid bases inside the cell is to overexpress Ypc 1p, an alkaline ceram idase that catalyzes the hydrolysis of phytoceramide to phytosphingosine (Mao et al., 2000). Overexpression of YPC1 also resulted in repression of pFET3-lacZ activity.

PAGE 97

97 Exogenous and Endogenous Sphingoi d Bases Mimic the Effect of IZ H2 Overexpression Both overexpression of Ypc1p and addition of phytosphingosine or dihydrosphingosine (Figures 4-4A and 4-4B) caused p FET-lacZ or ferroxidase repr ession that depends on PKH1 PKH2 TPK1 TPK2 TPK3 RAS2, SNF4 SIP1 S IP2 NRG1 and NRG2 Deletion of PKH1 PKH2 TPK1 TPK2 TPK3 NRG1 and NRG2 resulted in alleviation of p FET3-lacZ repression caused by a break in the signal tr ansduction pathway (Figures 4-5A 4-5E, and 4-5F), while only partial alleviation was demonstrated in the snf4 sip1 and sip2 mutant strains (Figure 4-5D). This data suggests that Ypc1p, phytosphingosin e, or dihydrosphingosine activates a similar pathway as PAQR overexpres sion (Kupchak et al., 2007). IZH2 Overexpression is Affecting Sphingoid Metabolism In this study, we postulated that IZH2 m ay require the de novo biosynthesis of sphingoid bases and ceramidases by affecting serine palmitoy ltransferase (SPT). To test this hypothesis, we used a SPT inhibitor, myriocin (Miyake et al., 1995) (Figure 6-6A). We showed that myriocin fully alleviated p FET3-lacZ repression of IZH2 overexpression and demonstrated maximal effect at 10 M (Figure 4-7A). But, myriocin did not alleviate p FET3-lacZ repression by YPC1 overexpression, phytosphingosine addition, or in iron-replete LIM (Figure 4-7B). Sphingoid bases can also accumulate by inhibiting ceramide synthases Lag1p and Lac1p, via the antifungal drug fumonisin B1 (FB1) (Wang et al., 1991). Thus, we examined the effect of fumonisin B1 on pFET3-lacZ activity (Figure 4-6B). p FET3-lacZ is partially repressed at 1 M FB1 and maximum repression occurs at 10 M FB1 (Figure 4-6C). We also tested the ability of two inhibitors of sphingosine kinase (SK), N,N-dimetylsphi ngosine (N,N-DMS) and sphingosine kinase inhibitor (SPH KI) to block the conversion of sphingosin e to sphingosine-1-phosphate (Figure 4-6C). p FET3-lacZ is fully repressed by the addition of 10 nM N,N-DMS, while SPH

PAGE 98

98 KI repressed p FET3-lacZ at 10 M in the BY4742 wild strain carrying the empty vector grown in iron-defecient LIM (Figure 4-7C). Accumulation of Sphingoid Bases Effects FET3 To further exam ine sphingoid base accumulati on, we investigated th e effects of deleting genes involved in complex sphingolipid formation. Therefore, we looked at p FET3-lacZ activity in strains lacking SCS7, SUR1 CSG2 IPT1 We show that scs7 sur1 csg2 and ipt1 mutant strains exhibited c onstitutive repression of p FET3-lacZ in iron-deficient LIM (Figure 48A). In this study, we also inve stigated the effect of deleting SUR2 which is responsible for converting dihydrosphingosine into phytosphi ngosine (Haak et al ., 1997). In the sur2 mutant strain p FET3-lacZ is constitutively repressed (Figure 4-8B). Also, we studied strains lacking, either, long chain fatty acid elongases, sur4 and fen1 or long chain base kinases, lcb4 and lcb5 These strains all exhibited constitutive repression of p FET3-lacZ activity; however, IZH2 overexpression did not affect p FET3-lacZ in these strains in iron-defi cient LIM (Figures 4-8C). Endogenous Alkaline Ceramidase are not Required for FET3 Signaling In this study we also investigated IZH2 ability to act as an alkaline ceramidase. We measured the effect of D-erythro-MAPP (Bielaws ka et al., 1996) (Figure 4-9), an inhibitor of YPC1 and YDC1 on Izh2p-dependent repression p FET3-lacZ repression. D-erythro-MAPP alleviated repression of p FET3-lacZ by IZH2 overexpression and exhibits its maximal effect at 100 nM in iron-deficient LIM. It also eliminated the repression caused by YPC1 overexpression on pFET3-lacZ ; however, D-erythro-MAPP did not have an effect on p FET3-lacZ repression when phytosphingosine was added to iron-deficient LIM or when the experiment was performed in iron-replete LIM (Figure 4-10A). We also studied the effect of IZH2 overexpression in a strain lacking YPC1 and YDC1 which are the endogenous alkaline ceramidases. Un expectedly, ferroxidase activity was partially

PAGE 99

99 repressed in this strain. Ne vertheless, Izh2p overexpression slightly repressed ferroxidase activity, while Ypc1p overexpression fully repressed ferroxidase activity in iron-defecient LIM (Figure 4-10B). We also examined the accumula tion of sphingoid bases by inhibiting serine palmitoyltransferase, ceramide synt hase, and sphingosine kinase in the ypc1 ydc1 mutant strain. Ferroxidase activity in the ypc1 ydc1 strain was partially alle viated by the addition of 100 nM Derythro -MAPP and 1 M myriocin, but not by 10 nM N,N-DMS in iron-deficient LIM (Figure 4-10C). Discussion The Izh pro teins in yeast belong to a large family of membrane receptors known as PAQRs, which also include related proteins fr om human and bacteria. The PAQR family is unified by the presence of seven transmembrane domains and three highly conserved, short, amino acid motifs (Lyons et al., 2004). The PAQRs al so share distant similarity with a family of proteins known as the alkaline ceramidases as shown in chapter 1. Therefore, the similarities between the PAQRs and the al kaline ceramidases suggest th ey both may possess the same enzymatic activity. Effect of Sphingoid Bases on FET3 Similar to the Izh2p, the overexpression of YPC1 an alkaline ceramidase that hydrolyzes phytoceramide to phytosphingosine, repressed FET3 This effect on FET3 could also be recapitulated by the addition of exogenous sphingoi d bases. In this case, sphingosine (SPH), phytosphingosine (PHS), and dihydr osphingosine (DHS) all repressed FET3. Also molecules that had a similar structure to the sphingoid ba ses, such as stearylamine, could not repress FET3 suggesting that the effect is sp ecific for sphingoid bases. Even though the addition of exogenous s phingoid bases can cause an effect on FET3 this does not mean that sphingoid bases and the Izh2p represses FET3 by identical mechanisms. In

PAGE 100

100 yeast, sphingoid bases are known to activate Pk h1p and Pkh2p. This is consistent with our hypothesis that sphingoid bases act as a second messenger in the repression of FET3 As with IZH overexpression, FET3 repression by exogenous sphingoid bases or by YPC1 overexpression also requires Pkh1p/Pkh2p, PKA, AMPK, and Nrg1p/Nrg2p. Thus suggesting, with minor exceptions, that Izh2p and sphingoid bases may activate a similar pathway in yeast. Izh2p-Dependent Signaling is Alleviated By Disruption of Sphin goid Base Synthesis The question arose of how Izh2p affects sphi ngoid base levels. The Izh2p could alter sphingoid base levels in two ways: 1) it could increase de novo biosynthesis by affecting serine palmityltransferase (SPT) or, 2) increase ca tabolism of ceramide by activating ceramidase activity. Therefore, we first tested IZH2 overexpression effect on de novo sphingoid base biosynthesis. De novo sphingoid base biosynthesis begins in the endoplasmic reticulum where SPT, which is composed of Lcb1p, Lcb2 and Tsc3p subunits, catalyzes the condensation of serine with fatty acid-CoA to yield 3-ketosphinganine (Hanada et al., 2003). Tsc10 catalyzes the reduction of 3-ketosphingani ne to dihydrosphingosine (Beeler et al., 1998), which is then hydroxylated by SUR2 to form PHS (Haak et al., 1997). To test if IZH2 overexpression effects de novo sphingoid base biosynthesis, we added the SPT inhibitor, myriocin. Treatment with my riocin caused alleviat ion of Izh2p-dependent repression of FET3 in a dose dependent manner. However, the addition of myriocin could not effect FET3 repression when YPC1 was overexpressed, or when we added PHS or iron to the media. This suggests myriocin does not cause a general effect on FET3 transcription. Accumulation of Sphingoid Bases Represses FET3 The fact that myriocin treatment a lleviates Izh2p-dependent repression of FET3 indicates signaling must require de novo sphingoid base biosynthesis. We therefore tested if de novo ceramide biosynthesis was also required. The ce ramide synthases, Lag1p and Lac1p, catalyze

PAGE 101

101 the formation of ceramidases from sphingoid base s (Guillas et al., 2001; Schorling et al., 2001). Lag1p and Lac1p are inhibited by fumonisin B1 (FB1), which caused the build-up of sphingoid bases. The addition of fum onisin to cells carrying the empt y vector exhibited constitutive repression of FET3 Thus IZH2 -dependent repression of FET3 is inhibited by myriocin, while FB1 constitutively activates it. This result suggests that sphingoid base accumulation may function downstream of Izh2p in this signal transduction pathway. Accumulaton of sphingoid bases can also be generated in a variety of other artificial conditions, and could be accomplished through in hibiting sphingoid base degradation. Excess sphingoid base is degraded by a pair of long chain base (LCB) kinases, Lcb4p and Lcb5p, which phosphorylate the sphingoid base to ma ke sphingoid base-1-phosphate. In S. cerevisiae, Lcb4p accounts for the majority of LCB kinase activit y (Nagiec et al., 1998). The newly formed phytosphingosine-1-phosphate (PHS-1-P) or dih ydrosphingosine-1-phosphate DHS-1-P can then be cleaved by an LCB phosphate lyase, Dpl 1p (Saba et al., 1997), to yield ethanolamine phosphate and a C16 aldehyde. We postulated that since sphingoid bases accumulate in strains lacking Lcb4p or Lcb5p, or when LCB kinases are inhibited that a decrease in sphingoid base phosphorylation would result in repression of FET3 First, we tested the effects of the LCB kinase inhibitors, N,N-dimethylsphingosine (N,N -DMS) and sphingosine kinase inhibitor 2 (SPH KI), and found that both N,N-DMS and SPH KI repressed FET3 in an Izh2p-independent manner. Second, in cells lacking LCB4 or LCB5 FET3 is also constitutively repressed. However, the loss of LCB4 has a more significant effect on FET3 than the loss of LCB5 suggesting LCB4 consistent with its role in sphingoid base phosphorylation. A second way to create the accumulation of PH S is to delete the genes involved in the synthesis of complex sphingolipids. Once a ceram ide is formed, it is transferred to the golgi

PAGE 102

102 apparatus, where it is hydroxylates the ceramide fatty acid by Scs7p (Dunn et al., 1998). Inositol phosphate is then placed onto the ceramide by Aur1p to form inositol phosphoceramide (IPC) (Haak et al., 1997). The enzyme complex Sur 1p, Csg2p, and Csh1p then catalyzes the formation of mannose-inositol-phosphoceramide (MIPC) from GDP-mannose (Beel er et al., 1997) (Uemura et al., 2003) (Lisman et al., 2004). The fo rmation of the most complex sphingolipid in yeast, mannose-(inositol-phosphate)2-ceramide (M(IP)2C), is made by the transfer of a second inositol phosphate to MIPC, a reac tion that requires Ipt1p (Dickson, et al., 1997). Therefore, we decided to test strains that were lacking these genes involved in complex sphingolipid biosynthesis except for the aur1 mutant strain, which is lethal (Nagiec et al., 1997). In the scs7 sur1 csg2 or ipt1 mutant strains, FET3 was constitutively repressed suggesting a reversal of this pathway and a build-up of sphingoid bases. Another way to create excess amounts of PHS is to delete SUR4 or FEN1 genes. Both of these genes are essential for the conversion of C16 fatty acids to very long chain fatty acids (C20 to C26) that are required for ceramides (Oh et al., 1997). Therefore, we tested the effect of knocking out SUR4 and FEN1 on FET3 In sur4 and fen1 mutant strains, FET3 exhibited constitutive repression. We were also interested if the accumula tion of excess endogenous DHS would have the same effect as the addition of exogenous DHS. Accumulation of DHS can occur by the deletion of the SUR2 gene, which is responsible for converting DHS into PHS (Haak et al., 1997). In the sur2 mutant strain, FET3 is constitutively repressed s uggesting that PHS and DHS have a similar effect on FET3 Inhibition of Ceramida se Activity Alleviate IZ H s and YPC1 Effect on FET3 The Izh2p could also alter sphingoid base leve ls by increasing catabolism of ceramide by activating ceramidase activity. Firs t, we measured the effect of Derythro -MAPP, an alkaline

PAGE 103

103 ceramidase inhibitor, on IZH2 -dependent repression of FET3 and found that D-erythro -MAPP was able to partially al leviate the repression of FET3 by both YPC1 and IZH2 overexpression. To verify our results that IZH overexpression was causing the accumulation of sphingoid bases, Nancy Villa studied accumulation of s phingoid bases by overexpressing Izh2p. In the experiment, she found a modest accumulation of both phytosphingosine and dihydrosphingosine compared to vector control. Moreover, this accumulation suggests that the Izh2p may be acting as ceramidases and that Izh2and Ypc1p-de pendent signaling requires the conversion of sphingoid bases to ceramides prior to the formation of the sp hingoid second messenger responsible for FET3 repression. FET3 Signa ling does not Require Endogenous Alkaline Ceramidases Previous results have shown myri ocin inhibits Izh2p-dependendent FET3 repression, but not Ypcp1por exogenous PH S-dependent repression of FET3 in a wild type strain. This suggests Izh2p requires newly synthesized ceramides for its effect on FET3, while Ypc1p does not. This was our first indication that Izh2p do es not work through Ypc1p. In order to study Ypc1p role in Izh2p signaling, we obtained a strain lacking YPC1 and YDC1 This strain was used because it has no endogenous ceramidase activity. Therefore, we tested if Izh2p could repress FET3 in ypc1 ydc1 mutant strain. Strikingly, FET3 was partially repressed in this strain; however, Izh2p overexpression did slightly enhance FET3 repression. This suggests Izh2p-dependent signaling is independent of alkaline ceramidases. The low expression of FET3 in the ypc1 ydc1 mutant strain can be explained by a decr ease in ceramide turnover resulting in an increase of PHS/DHS accumula tion by feedback inhibition. This result is supported by myriocin alleviating the c onstitutive repression of FET3 while N,N-DMS further represses FET3 (Figure 4-11).

PAGE 104

104 Summation The PAQR fam ily shares similarities with the alkaline ceramidases. This led us to postulate that the Izh proteins may be acting as ligand-regulated ceramidases that control the balance between ceramides and sphingoid bases. Ou r data supports two main conclusions. First, repression of FET3 by IZH -dependent signaling requires de novo ceramide biosynthesis and that this signal transduction on FET3 can be stopped by inhibition of ceramidase activity. Second, sphingoid bases acts a second messenger and signals FET3 repression via a similar mechanism as seen by the yeast and human PAQRs.

PAGE 105

105 Figure 4-1. Phylogenetic analysis of PAQRs. A bootstrapped phylogene tic tree is shown the relationship between the PAQRs from various sources. Izh1p and Izh2p cluster with AdipoR1 and AdipoR2. Yeast and human GPCRs form a distinct clade and are included as an outgroup, which the tree is root ed. The length of the tree branches is proportional to the calculated distance between sequences with the scale bar indicating 0.1 substitutions per site. Numbers at the nodes are confidence values that refer to the number of times per 1,000 tr ees drawn a particular grouping is made.

PAGE 106

106 Figure 4-2. Topology predicted for all members of the PAQRs and alkaline ceramidases with the locations and the consensus sequences of the three highl y conserved motifs. Grey represents membrane, ovals indicate tran smebrane domains, and circles represent conserved regions of PAQR family. Class I, III PAQRs, and alkaline ceramidases are presumed to have 7 transmembrane doma ins while Class II PAQRs may have 8 transmembrane domains.

PAGE 107

107 inactive (cyto)Nrg1p/Nrg2p Nrg1p/Nrg2p active inactive AMPK AMPKactiveMsn2p/Msn4p Msn2p/Msn4pactive active active inactivePKA PKA active inactivePkh1p/Pkh2pPkh1p/Pkh2p ?Phytoceramide Phytosphingosine Ligand (cyto) (nucl) (nucl) (nucl) (cyto) (cyto) (nucl) CCCTC FET3 Figure 4-3. A model for the mechanism of FET3 by IZH2 overexpression. Solid arrows indicate protein activation or localization of pr otein. Broken arrows indicate protein interaction on localization a nd activation. Cyto = protein is cytoplamic. Nucl = protein is nuclear. Shaded blue portions represent areas studied in chapter.

PAGE 108

108 0 -Galactosidase Activity % of WT Low Fe3+ Vector Control 0 20 40 60 80 100 120 140 Sphingosine Phytosphingosine Dihydrosphingosine Stearylamine 0.1 1.0 10 100 M [LCB]A Figure 4-4. Exogenous and endoge nous sphingoid bases affect FET3 BY4742 wild type strain is used in panels A and B. -Galactosidase activities are shown as a percentage of fully induced -galactosidase in wild type strain carrying an empty control vector grown in iron-deficient or iron-replete LIM. The error bars represent 1 standard deviation for experiments performed in triplicate. (A) -Galactosidase activities of the p FET3-lacZ reporter are induced in iron-deficient LIM in the wild type strain when sphingosine, phytosphingosine, or di hydrosphingosine was added, but not with stearylamine. (B) YPC1 overexpression exhibited an effect on the p FET3-lacZ reporter in iron-deficient LIM.

PAGE 109

109 Control -Galactosidase Activity % of WT Low Fe3+ Vector Control 0 20 40 60 80 100 120 1 M Fe3+ 1 mM Fe3+ Ypc1pB Figure 4-4 continued.

PAGE 110

110 -Galactosidase Activity % of WT Low Fe3+ Vector Control 0 20 40 60 80 100 120 140 160 Control Ypc1p PHS DHS Wild Type nrg1nrg2A Figure 4-5. Sphingoid bases cause an IZH -dependent repression of FET3 BY4742 wild type strain is used in panels A to D. -Galactosidase activities ar e shown as a percentage of fully induced -galactosidase activity in wild type strain carrying an empty control vector grown in iron-deficient or iron-re plete LIM. The error bars represent 1 standard deviation for experiments pe rformed in triplicate. (A) Loss of NRG1 and NRG2 exhibit loss of repression of p FET3-lacZ (B) -Galactosidase activities of the pFet3-lacZ reporter are represse d in strains lacking GAL83 Ypc1p overexpression partially alleviated pFET3-lacZ repression. Strains lacking AMPK (Snf4p) or (Sip1p and Sip2p)-subunits partially alleviated p FET3-lacZ repression when Ypc1p was overexpressed or with the addition of phytosphingosine or dihydrosphingosine. (C) -Galactosidase activities of the p Fet3-lacZ reporter are alleviated in strains lacking TPK1 TPK2 and TPK3 but only partially alleviated in the RAS2 deleted strain when Ypc1p was overexpressed or phytosphingosine or dihydrosphingosine was added to iron-deficient LIM. (D) -Galactosidase activities of the p FET3-lacZ reporter are alleviated in strains lacking PKH1 and PKH2

PAGE 111

111 -Galactosidase Activity % of WT Low Fe3+ Vector Control 0 20 40 60 80 100 120 140 Control Ypc1p PHS DHS Wild Type sip1sip2snf4B Wild Type -Galactosidase Activity % of WT Low Fe3+ Vector Control 0 20 40 60 80 100 120 140 160 Control Ypc1p PHS DHS ras2tpk1tpk2tpk3C Figure 4-5 continued.

PAGE 112

112 -Galactosidase Activity % of WT Low Fe3+ Vector Control 0 20 40 60 80 100 120 140 160 Control Ypc1p PHS DHS Wild Type pkh1pkh2D Figure 4-5 continued.

PAGE 113

113 A O NH2O HOH O O OH OH N N NN NH2P O-O O P O O-O CH3CH3OH O NH O NH S O CH3 CH3O NH2OH + CH3OH OH NH2O OH OH Figure 4-6. Inhibitors of sphi ngolipid synthesis and breakdown. (A) Myriocin inhibits the formation of 3-ketodihydrosphingosine fr om palmityl-CoA and L-serine. (B) Fumonisin B1 inhibits the formation of phytocer amide from phytosphingosine. (C) N,N-dimethylsphingosine or sphingosine kinase inhibitor 1 inhibits the formation of dihydrosphingosine-1-phosphate or phyt osphingosine-1-phosphate from dihydrosphingosine or phytosphingosine. Dihydrosphingosine-1-phosphate or phytosphingosine-1-phosphate is broken down to palmityl-aldehyde and phosphorylethanolamine.

PAGE 114

114 B CH3NH2OH OH OH CH3NH OH OH OH R O CH3CH3OH NH2OHOH O O O HO OH O CH3CH3O OO OH OOH C CH3NH2OH OH OH CH3NH2O OH OH P OH OH O CH3N OH OH OH CH3CH3 Cl S N NH OH OR Figure 4-6 continued.

PAGE 115

115 -Galactosidase Activity % of WT Low Fe3+ Vector Control 0 20 40 60 80 100 120 Control Izh2p Ypc1p PHS Control (1 mM Fe3+) 0 0.11.0 10100 M [Myriocin]A Figure 4-7. IZH overexpression is affecting sphingoid me tabolism. BY4742 wild type strain is used in panels A to C. -Galactosidase (Panel A to C) activities are shown as a percentage of fully induced -galactosidase activity in w ild type strain carrying an empty control vector grown in iron-deficien t or iron-replete LIM. The error bars represent 1 standard deviation for experi ments performed in tr iplicate (A) Myriocin alleviates repression of p FET3 -lacZ in BY4742 wild type strain carrying the IZH2 overexpressor, but not when YPC1 is overexpressed or phytosphingosine is added. (B) Myriocin alleviates repression of p FET3-lacZ due to the activation of Izh2p by thaumatin in 0.05% galactose iron-defici ent LIM. (C) N,N-dimethylsphingosine (N,N-DMS), fumonisin B1 (FB1), and s phingosine kinase inhibitor 1 (SPH KI) repressed p FET3-lacZ in the BY4742 wild type strain carrying the empty vector but was not affected by IZH2 overexpression.

PAGE 116

116 -Galactosidase Activity % of WT Low Fe3+ Vector Control 0 20 40 60 80 100 120 140 Control (2% Galactose) Izh2p (2% Galactose) Control (0.05% Galactose) Izh2p (0.05% Galactose) Control (1 mM Fe3+) 0 0.11.0 10100 M [Myriocin]B Figure 4-7 continued

PAGE 117

117 -Galactosidase Activity % of WT Low Fe3+ Vector Control 0 20 40 60 80 100 120 Control + FB1 Izh2p + FB1 Control + N,N-DMS Izh2p + N,N-DMS Control + SPH KI Izh2p + SPH KI 0 0.1 1.0 10 100 nM [Inhibitor] 103104C Figure 4-7 continued

PAGE 118

118 -Galactosidase Activity % of WT Low Fe3+ Vector Control 0 20 40 60 80 100 120 Control Izh2p Wild Type scs7sur1csg2ipt1A Figure 4-8. Accumulation of sphingoid bases effects FET3 BY4742 wild type strain is used in panels A to C. -Galactosidase activities are shown as a percentage of fully induced -galactosidase activity in wild type strain carrying an em pty control vector grown in iron-deficient or iron-replete LIM. The erro r bars represent 1 st andard deviation for experiments performed in trip licate. (A) Deletion of SCS7, SUR1 CSG2 and IPT1 carrying the empty vector constitutively repressed pFET3-lacZ (B) Deletion of SUR2 carrying the empty vector constitutively repressed pFET3-lacZ (C) Deletion of LCB4 LCB5 SUR4 and FEN1 carrying the empty vector constitutively repressed pFET3-lacZ

PAGE 119

119 Wild Type -Galactosidase Activity % of WT Low Fe3+ Vector Control 0 20 40 60 80 100 120 Control Izh2p sur2B -Galactosidase Activity % of WT Low Fe3+ Vector Control 0 20 40 60 80 100 120 Control Izh2p Elongases Wild Type sur4fen1lcb4lcb5 LCB KinasesC Figure 4-8 continued.

PAGE 120

120 OH CH3NH CH3 CH3NH OH OH OH R O CH3NH2OH OH OH Figure 4-9. Inhibitor of ceramid e breakdown. (A) D-erythro-MA PP inhibits the breakdown of phytoceramide to phytosphingosine.

PAGE 121

121 0 -Galactosidase Activity % of WT Low Fe3+ Vector Control 0 20 40 60 80 100 120 Control Izh2p Ypc1p PHS Control (1 M Fe3+) 0.1 1.0 10100200 nM [D-erythro-MAPP]A Figure 4-10. Involvement of alkaline ceramid ases in Izh2p-dependent signaling. BY4742 wild type strain is used in panels A to C, while ypc1 ydc1 was used in panels B and C. -Galactosidase (Panel A) and ferroxidase (Panel B and C) activ ities are shown as a percentage of fully induced -galactosidase and ferroxidase activity in wild type strain carrying an empty cont rol vector grown in iron-defi cient or iron-replete LIM. The error bars represent 1 standard deviation for experiments performed in triplicate. (A) Derythro-MAPP (D-MAPP) alleviates repression of pFET3-lacZ in BY4742 wild type strain carrying the IZH2 or YPC1 overexpressor, but not when phytosphingosine is added. (B) Deletion of YPC1YDC1 carrying the empty vector causes partial repression of ferroxidase activity, while IZH2 overexpression only causes a slight repression in this strain. Ypc1p overexpression restored repression of pFET3-lacZ in the ypc ydc mutant strain (C) Ferroxidase activity in ypc1 ydc1 mutant strain carrying the empty vector is alleviated by the addition of Derythro MAPP or myriocin, but is not effected by N,N-DMS.

PAGE 122

122 Wild Type Ferroxidase Activity % of WT Low Fe3+ Vector Control 0 20 40 60 80 100 120 Control Izh2p Ypc1p ypc1ydc1B Ferroxidase Activity % of WT Low Fe3+ Vector Control 0 20 40 60 80 100 120 140 Control D-MAPP Myriocin N,N-DMS Wild Type ypc1ydc1C Figure 4-10 continued.

PAGE 123

123 Figure 4-11. Role of sphingolipids in IZH -dependent signaling of FET3. Sphingolipids are made in the endoplasmic reticulum (ER) and migrates to the plasma membrane (PM) where they are involved in a variety of roles in yeast. Complex sphingolipids function on the cell surface while ceramida ses and sphingoid bases have roles in signal transduction. Excess sphingolipids are degraded by a catabolic pathway. Ovals indicate the known sites of alkaline ceramidases activity a nd proposed sites of Izh2p action. Boxes indicate sites of inhibitor action. Dotted arrows indicate mechanisms by which elevation of sphingoid base may result. CAPP = C eramide A ctivated P rotein P hosphatase. In yeast nomenclature a gene IZH2 encodes a protein called Izh2p can be deleted to make a knockout strain referred to as izh2

PAGE 124

124 CHAPTER 5 PAQRS REPRESS FET3 IN SIMIL AR MANNER AS IZH2 Introduction The reason f or this study is data indicating that integral membrane protein receptors, which belong to the recently discovered PAQR family of receptors, are involved in regulating high affinity iron-uptake, which belong to the rece ntly discovered PAQR family of receptors. The human PAQR family consists of eleven prot eins, which can be grouped into three classes based on sequence similarity and structural features. The human genome encodes four Class I receptors (AdipoR1, AdipoR2, PAQR3, and PA QR4), which are distinguished by seven transmembrane domains. Two of these proteins are adiponectin receptors, AdipoR1 and AdipoR2, which bind to its ligand adiponectin. Adiponectin is a hormone that has an insulinsensitizing anti-diabetic effect and is a poten tial link between obesity and type II diabetes (Yamauchi et al., 2003). AdipoR1 and AdipoR2 have a confirmed topology with the C-terminus found on the outside of the plasma membrane (K im et al., 2003; Yamauchi et al., 2003). The Class II clade consists of five proteins (mPR mPR mPR PAQR6, and PAQR9) (Zhu et al., 2003a). Three of these proteins, mPR mPR and mPR bind to the steroid hormone, progesterone, which has many roles in reproduction, sexual behavior, ovulation and pregnancy (Mulac-Jericevic et al., 2000; Conneely et al., 2002; Mulac-Jericevic and Conneely, 2004). Due to its numerous functions, progesterone and its related derivatives have an array of medical uses ranging from birth control to breas t cancer treatment (Sitruk-Ware and Plu-Bureau, 1999). Unlike Class I recepto rs, Class II receptors are predicted to have an eighth transmembrane domain on the C-terminus. Two more proteins in this family have also been identified, PAQR10 and PAQR11, which are highly divergent and are more cl osely related to the bacterial hemolysin III than the other PAQRs (Chen et al., 2004). These proteins make up the

PAGE 125

125 Class III PAQRs. The presense of three highly conserved motifs, which are explained in chapter 1, are found on the cytoplasmic side of the membrane enables us to unify the three classes of human PAQRs (Lyons et al., 2004). Saccharomyces cerevisiae also possesses four Class I PAQR receptors named Izh1p-4p (Lyons et al., 2004). One of these receptors, Izh 2p, is a receptor for osmotin, a plant defensin, that is structurally similar to adiponectin. Osmotin also has be en shown to activate adiponectin receptors, which suggests there is conservation of function (Narasimhan et al, 2005), Therefore, yeast is an excellent model for th e study of PAQR receptors. We chose yeast as a model system for three reas ons: 1) Yeast is a eukaryotic system and its simplicity allows the user to elucidate a variety of basic biologi cal processes. 2) S. cerevisiae contain receptors in the PAQR family suggestin g that the parts require d to interpet second messenger signals produced by these proteins is pr esent (Lyons et al., 2004). 3) Yeast does not produce adiponectin or progesterone, and therefore, do not have have receptors for these ligands. This feature allows the user to investigate the ac tivity of a specific human receptor of interest. We show that the PAQRs result in repression, in most cases, of the gene encoding FET3 To examine this principle, we used -galactosidase and ferroxidase assays to demonstrate that PAQR overexpression from the three cl asses results in repression of the FET3 gene under conditions of iron deficiency in the same manner as Izh2p. However, AdipoR2 and mPR were not capable of signaling repressi on when overexpressed. However, if their respective activating ligands were added to the growth medium, (adiponectin for AdipoR2 or progesterone for mPR ) repression of FET3 was observed. In this study, we show that the human PAQRs regulate the Nrg1p/Nrg2p repressors. Furthermore, we demonstrate that the human PAQRs, like the yeast PAQRs, require cAMP

PAGE 126

126 dependent kinase (protein kina se A, PKA), AMP-dependent ki nase (AMPK), and the sphingoid base-sensing kinases Pkh1p and Pkh2p (Kupchak et al., 2007). Thus we report that the PAQRs demonstrate a negative effect on FET3 and this effect occurs using a similar mechanism in which the yeast Izh2p represses FET3 Materials and Methods Yeast Strains and Plasmids W ild type (BY4742) and mutant strains in this background (pkh1, pkh2, ras2, tpk1, tpk2, tpk3, snf1, snf4, gal83, sip1, sip2, nrg1, nrg2, msn2, and msn4) were purchased from Euroscarf ( http://web.uni-frankfu rt.de/fb15/m ikro/euroscarf/col_index.html ) and their genotypes are listed in Supplementary Table 1. The p FET3-lacZ plasmid (Gift of D. Eide) has the promoter region of the FET3 gene fused to the lacZ open reading frame and is URA3 selectable. PAQR prot eins (PAQR3, PAQR4, mPR mPR and PAQR11) were cloned from cDNAs into the LEU2 selectable pRS316GAL1 plasmid (Lyons et al., 2004). The plasmid uses the GAL1 promoter to drive galactose-inducible expr ession of cloned genes. The cDNAs used as templates were obtained from Open Biosystems or the NEDO Human cDNA Sequencing Project and primers used to clone PAQR3, PAQR4, mPR mPR and PAQR11 are listed in Supplementary Table 2. The cDNA for PAQR3 had one intron that we removed by overlap extension PCR. All clones were sequenced to ensure that no errors occurred during PCR. Galactosidase and Ferroxidase Assays -Galactosidase and ferroxidase sssays were perfor med as described in chapter 2 except for the addition of the following compounds to LIM. Progesterone (Sigma #P8783-1G), 17 hydroxyprogesterone (#H572-5G), 21-hydroxyprogesterone (Sigma #D6875-500MG), 17 21-

PAGE 127

127 hydroxyprogesterone (Sigma #R0500-1G) cortisol and testosterone (Sigma #T1500-5G) were added to the cultures at va ried concentrations. Results PAQRs are found in all eukaryotes ranging from Saccharomyces cerevisiae to Homo sapiens. Human PAQRs are composed of both adiponectin receptors and membrane progestin receptors. Overexpression of both yeast and human PAQRs causes repression of FET3 Thus, experiments were designed to test if yeast a nd human PAQRs signal in a similar manner to the Izh2p. Effect of Human PAQRs on FET3 Expression Previous studies have shown that IZH2 overexpression negatively affected the transcription of the FET3 gene (Kupchak et al., 2007). Using a pFET3-lacZ reporter, we show that overexpression of Izh1p, Izh3p, Izh4p, AdipoR1, PAQR3, PAQR4, mPR mPR PAQR11, but not AdipoR2 and mPR results in repression of the p FET3-lacZ reporter construct even in iron-limiting media. To verify these results, we also measured the enzymatic activity of Fet3p to oxidize Fe2+ to Fe3+ using a ferroxidase assay (Figures 5-1A to 5-1C). AdipoR2 and mPR are Activated by its Putative Ligan ds. As was shown in Figures 5-1A and 5-1B, the overexpression of the Class I receptor AdipoR2 and the Class II receptor mPR were unable to repress p FET3-lacZ In this study, we demonstrate that the addition of adiponectin (Figure 5-1C) for AdipoR2 and progesterone for mPR to the iron-limiting LIM was effective at signaling p FET3-lacZ repression. Other steroids, such as 17 -hydroxyprogesterone, 21-hydroxyprogesterone, 17 21dihydroxyprogesterone during mPR overexpression were also able repress p FET3-lacZ for mPR but at a concentration 10-times higher than progesterone. The steroids cortisol and testosterone had no effect on p FET3-lacZ activity when mPR was overexpressed (Figure 5-1D).

PAGE 128

128 PAQRs affect FET3 Expression In chapter 2, we de monstrated that IZH2 overexpression altered FET3 regulation by affecting the transcriptio nal factors Msn2p/Msn4p and Nr g1p/Nrg2p. In this study, we demonstrated that Nrg1p and Nrg2p are also necessary for signal transduction of FET3 by the yeast and human PAQRs. PAQR dependent p FET3-lacZ repression was lost in nrg1 and nrg2 mutant strains (Figures 5-3A to 5-3C). PKA is known to function downstream of Izh2p (Kupchak et al., 2007). Therefore, we tested its role in PAQR-dependent repression of FET3 pFET3-lacZ activity exhibited loss of repression during PAQR overexpression in mutants lacking each of the th ree catalytic subunits, tpk1, tpk2, and tpk3 (Figures 5-5A to 5-5C). Upstream of PKA is Ras2p-cAMP, which is known to activate PKA (Jiang et al., 1998). In the ras2 mutant strain, PAQR overexpression did not cause repression of pFET3-lacZ (Figures 5-5A to 5-5C). Also downstream of Izh2p is the AMPK co mplex (Kupchak et al., 2007). AMPK is composed of subunits that control its activity and lo calization. Deletion of SNF4 the stimulatory -subunit of AMPK caused pa rtial repression of p FET3-lacZ while overexpression of the PAQRs in the snf4 mutant strain had no further effect. Strains lacking AMPKs cytoplasmic subunits, Sip1p and Sip2p, showed partial repression of p FET3-lacZ and also was not effected by overexpression of the PAQRs (Figures 5-3A and 5-3C). Besides AMPK and subunits, there are other proteins that intereac t with AMPK in yeast. One of these proteins, Sip3p, is also essential for PAQR dependent re pression as there loss results in the inability to repress p FET3lacZ (Figures 5-4A to 5-4C). In this study, we show the genes PKH1 and PKH2 are involved PAQR-dependent FET3 repression. Loss of Pkh1p and Pkh2p results in complete loss of p FET3-lacZ repression.

PAGE 129

129 Although Pkh2p plays a greater ro le for repression when AdipoR1, AdipoR2, PAQR3, or PAQR4 are overexpressed; however, Pkh1p seems to be more important for the remaining PAQRs in repressing pFET3-lacZ activity (Figures 5-6A and 5-6B). Discussion The PAQR receptors are of critical importance to human he alth. First, AdipoR1 and AdipoR2 function as receptors for adiponectin, a protein secreted by adipocytes that has emerged as a candidate for the molecular link between obesity and type II di abetes (Kadowaki and Yamauchi, 2005). Second, human PAQRs also encode proteins for membrane progestin receptors, which are known to bind to progesterone (Zhu et al., 2003b). Finally, human PAQR10 and PAQR11 proteins make up the final class of receptors, which are presumed to be involved in monocyte to macrophage maturation (Bauer, et al., 2004). Human PAQRs were expressed in the eukaryotic model organism, Saccharomyces cerevisiae. An important reason for this choice is that yeast do not possess adiponectin or membrane progestin receptors, thus minimizing interference with data analysis due to progesterone or adiponectins inabilit y to bind to other PAQR receptors. PAQRs Repress FET3 Expression In previous studies, we dem onstrated that Izh2p activity could be monitored by using a galactosidase and ferroxidase assa ys, and that overexpression of IZH2 represses FET3 The Izh proteins belong to a larger fam ily of membrane proteins called the PAQRs, and consist of human adiponectin and membrane progestin receptors. Like Izh2p, overexpression of the yeast (Izh1p, Izh3p, and Izh4p) and human (AdipoR1, PAQR3, and PAQR4) Class I homologues and the human Class III homologue, (PAQR11) caused FET3 repression. AdipoR2 was unable to repress FET3 unless adiponectin was adde d to the media. Control experiments exhibited that adiponectin did not affect FET3 in yeast carrying the empty v ector, thereby indicating that

PAGE 130

130 adiponectins effect was specific for Adi poR2. Thus, human PAQR homologues can be functionally expressed in yeast. We propose that some human PAQR homologues, like many signaling proteins, have intrinsi c basal signal capability that is amplified by overexpression, making the presence of an activating ligand unn ecessary. AdipoR2, on the other hand, requires adiponectin for activation to signal the repression of FET3 Effect of Class II Homologues on FET3 Expression Hu man Class II PAQRs, mPR mPR and mPR were expressed in yeast and our experiments demonstrated that mPR and mPR were capable of repressing FET3 while mPR did not signal repression. mPR was activated by the addition of progesterone to transmit the repression signal. The effect of steroids on mPR is highly specific for progesterone: 17 hydroxyprogesterone, 21-hydr oxyprogesterone, and 17 ,21-dihydroxyprogesterone were an order of magnitude less effective than progesterone at activation, wh ile cortisol and testosterone were completely ineffective. Control experime nts showed that the addi tion of progesterone did not affect FET3 in yeast that were carrying an empty vector, indicating that progesterone does not non-specifically affect FET3 Therefore, we conclude that mPR and mPR are capable of basal signal transduction, while mPR is stimulated by progesterone to affect FET3 PAQRs Repress FET3 by a Common Mechanism All members of the human PAQRs either ha ve basal signaling capability or can be activated to repress FET3 Therefore, we propose that the PAQRs have a common signaling mechanism upstream of FET3 This possibility is supported by several lines of evidence from the literature. First, both adiponectin (Xi et al., 2005) and progesterone (Alkhalaf et al., 2002) have been reported to activate PDK (Vanhaeseb roeck and Alessi, 2000). The requirement for progesterone can be seen in th e re-initiation of meiosis in Xenopus oocytes, an event that requires mPRs, and can be bypassed by the addition of the PDK activator sphingosine (Strum et al.,

PAGE 131

131 1995). Also, PDK has been shown to play an im portant part in insuli n signaling (Mora et al., 2005), a process that requires adiponectin. Therefore we investigated if Pkh1p and Pkh2p were required for PAQR-dependent repression of FET3 Similar to Izh2p, the PAQRs require both Pkh1p and Pkh2p to repress FET3 Previous studies have also shown the e ffects of adiponectin and progesterone are alleviated by inhibiting PKA activity (Kline and Karpinski, 2005) (Ouchi et al., 2000). Therefore, we investigated if the PKAs subun its were necessary for signal transduction of FET3 repression. Indeed, TPK1 TPK2 TPK3 and RAS2 were all needed for FET3 repression to occur. Based on previous reports in our laborator y, we also demonstated the AMPK complex was necessary for signal transduction. Our resu lts were verified by studies in which human AdipoR1 signals downstream through AMPK (Yamauchi et al., 2003). Our studies also demonstrated that FET3 repression requires the AMPK -subunits (Sip1p, and Sip2p), and subunit (Snf4p) as shown in chapter 3. Also, a cytolasmic Snf1p inter acting protein, Sip3p, is required for repression (Lesage, et al., 1994). This is intri guing since its human homologue, APPL1, is required for signal transduc tion by AdipoR1 (Mao et al., 2006). Even though the effect on FET3 by yeast PAQR requires Nr g1p/Nrg2p for repression and Msn2p/Msn4p for activation (Kupchak et al., 2007), there was no evidence in the literature to claim these transcription factors were re quired for PAQR-depe ndent repression of FET3 We proceeded to test hum an PAQRs effect on FET3 in knockout strains of both Nrg1p/Nrg2p, and indeed, the effect of the PAQRs on FET3 did require both Nrg1p/Nrg2p. Thus, FET3 repression by the human PAQRs requires Pkh1p, Pkh2p, Ras2p, Tpk1p, Tpk2p, Tpk3p, Sip1p, Sip2p, Snf1p, Snf4p, Nr g1p, Nrg2p, Msn2p, and Msn4p in yeast (Kupchak et al., 2007) and that ad iponectin and progesterone have been shown to regulate some

PAGE 132

132 of these proteins in vertebrate cells. This does not necessarily mean an identical signaling mechanism exists in vertabrates; however, our data support a model where the human PAQRs repress FET3 in yeast through a common mechanism. Summation Our findings support several im portant conclusions. First, every human PAQR tested, regardless of activating ligand and physiological function affected FET3 Second, we have presented data exhibiting that AdipoR2 and mPR senses and responds to its particular ligand when expressed in yeast cells. Third, the human PAQRs signal FET3 repression in a similar mechanism as the yeast PAQRs.

PAGE 133

133 Activity % of WT Low Fe3+ Vector Control 0 20 40 60 80 100 120 1 M Fe3+ 1 mM Fe3+ lacZlacZ fox fox Control Izh1p lacZlacZ fox fox Izh3p Izh4pA Figure 5-1. Class I, II, and III PAQRs repress FET3 transcription : BY4742 wild type strain is used in panels A, B and C. -Galactosidase (Panels A, B, and C) and ferroxidase (Panel A) activities ar e shown as a percentage of fully induced activity in wild type strain carrying an empty cont rol vector grown in iron-defi cient and iron-replete LIM. The error bars represent 1 standard deviation for experiments performed in triplicate. (B) -Galactosidase activity of the p FET3-lacZ reporter and the cell surface ferroxidase activity are induced in ir on-deficient LIM in th e wild type strain carrying the AdipoR1, PAQR3, PAQR4, mPR mPR and PAQR11 overexpressor, but not the AdipoR2 and mPR overexpressor. (D) AdipoR2 expression in 2% galactose has no effect on p FET3-lacZ unless the ligand adiponectin is added to irondeficient LIM. (E) mPR expression in 2% galactose has no effect on p FET3-lacZ unless a ligand is added to iron-defi cient LIM. Progesterone (PG), 17 hydroxyprogesterone (17 -HPG), 21-hydroxyprogesterone (21-HPG), and 17 ,21hydroxyprogesterone (17 ,21-DHPG) exhibited a repression of p FET3-lacZ while cortisol and testosterone (Test) had no effect on p FET3-lacZ repression.

PAGE 134

134 lacZ Activity % of WT Low Fe3+ Vector Control 0 20 40 60 80 100 120 140 1 M Fe3+ 1 mM Fe3+ lacZ lacZlacZ lacZ foxfoxfox foxfox Control AdipoR1AdipoR2 PAQR3PAQR4A Activity % of WT Low Fe3+ Vector Control 0 20 40 60 80 100 120 1 M Fe3+ 1 mM Fe3+ lacZ lacZ lacZlacZ lacZ foxfoxfox foxfox Control mPR mPR mPR PAQR11B Figure 5-1 continued.

PAGE 135

135 0 Galactosidase Activity % of Activity w/o Treatment 0 20 40 60 80 100 120 Control PAQR2 1.0 10100 [Adiponectin] pMC -Galactosidase Activity % of WT Low Fe3+ Vector Control 0 20 40 60 80 100 120 Control with PG PAQR7 with PG Control with 17 HPG PAQR7 with 17 HPG Control with 21-HPG PAQR7 with 21-HPG Control with 17 ,21-DHPG PAQR7 with 17 ,21-DHPG Control with Cortisol PAQR7 with Cortisol Control with Test PAQR7 with Test 0.0 1.010 100 1,000 nM [ligand]D Figure 5-1 continued.

PAGE 136

136 -Galactosidase Activity % of WT Low Fe3+ Vector Control 0 20 40 60 80 100 120 140 160 Control Izh1p Izh3p Izh4p Wild Type nrg1nrg2A Figure 5-2. Nrg1p and Nrg2p e ffect on PAQR-dependent FET3 repression: BY4742 wild type, nrg1 and nrg2 strains are used in panels A and B. -Galactosidase activities are shown as a percentage of fully induced activ ity in wild type stra in carrying an empty control vector grown in iron-deficient and ir on-replete LIM. The error bars represent 1 standard deviation for experiments performed in triplicate. (A) The repression of -galactosidase activities of the p FET3-lacZ reporter by overexpression of Izh1p, Izh3p, and Izh4p were absent in the nrg1 and nrg2 mutant strains. (B) The repression of -galactosidase activities of the p FET3-lacZ reporter by overexpression of AdipoR1, AdipoR2 with adiponectin, PAQR3, and PAQR4 were absent in the nrg1 and nrg2 mutant strains. (C) The repression of -galactosidase activities of the p FET3-lacZ reporter by overexpression of mPR with progesterone, mPR mPR and PAQR11 were absent in the nrg1 and nrg2 mutant strains.

PAGE 137

137 Wild Type -Galactosidase Activity % of WT Low Fe3+ Vector Control 0 20 40 60 80 100 120 140 160 Control AdipoR1 AdipoR2 PAQR3 PAQR4 nrg1nrg2B -Galactosidase Activity % of WT Low Fe3+ Vector Control 0 20 40 60 80 100 120 140 160 Control mPR mPR mPR PAQR11 CWild Type nrg1nrg2 Figure 5-2 continued.

PAGE 138

138 -Galactosidase Activity % of WT Low Fe3+ Vector Control 0 20 40 60 80 100 120 Control Izh1p Izh3p Izh4p AWild Type sip1sip2snf4 Figure 5-3: AMPK in FET3 repression: BY4742 wild type strain is used in panels A, B, C and D. snf4 sip1 and sip2 strains are used in panels A to C. -Galactosidase activities are shown as a percentage of fully induced activity in wild type strain carrying an empty control vector grown in iron-deficient and iron-replete LIM. The error bars represent 1 standard deviation for experiments performed in triplicate. (A to C) -galactosidase activities of the p FET3-lacZ reporter is partially repressed in snf4 sip1 and sip2 mutant strains and PAQR overexpression has no effect on activity.

PAGE 139

139 -Galactosidase Activity % of WT Low Fe3+ Vector Control 0 20 40 60 80 100 120 Control AdipoR1 AdipoR2 PAQR3 PAQR4 Wild Type sip1sip2snf4B -Galactosidase Activity % of WT Low Fe3+ Vector Control 0 20 40 60 80 100 120 Control mPR mPR mPR PAQR11 Wild Type sip1sip2snf4C Figure 5-3 continued.

PAGE 140

140 -Galactosidase Activity % of WT Low Fe3+ Vector Control 0 20 40 60 80 100 120 Control Izh1p Izh3p Izh4p Wild Type sip3A Figure 5-4. Sip3p in FET3 repression: BY4742 wild type strain is used in panels A, B and C. sip3 strain is used in panels A to C. -Galactosidase activities are shown as a percentage of fully induced activity in w ild type strain carry ing an empty control vector grown in iron-deficient and iron-re plete LIM. The error bars represent 1 standard deviation for experiments performe d in triplicate. (A) The repression of galactosidase activities of the p FET3-lacZ reporter by overexpression of Izh1p, Izh3p, and Izh4p were absent in the sip3 mutant strain. (B) The repression of galactosidase activities of the p FET3-lacZ reporter by overexpression of AdipoR1, AdipoR2 with adiponectin, PAQR3, and PAQR4 were absent in the sip3 mutant strain. (C) Th e repression of -galactosidase activities of the p FET3-lacZ reporter by overexpression of mPR with progesterone, mPR mPR and PAQR11 were absent in the sip3 mutant strain.

PAGE 141

141 -Galactosidase Activity % of WT Low Fe3+ Vector Control 0 20 40 60 80 100 120 Control AdipoR1 AdipoR2 PAQR3 PAQR4 Wild Type sip3B -Galactosidase Activity % of WT Low Fe3+ Vector Control 0 20 40 60 80 100 120 Control mPR mPR mPR PAQR11 CWild Type sip3 Figure 5-4 continued.

PAGE 142

142 -Galactosidase Activity % of WT Low Fe3+ Vector Control 0 20 40 60 80 100 120 140 160 Control Izh1p Izh3p Izh4p AWild Type ras2tpk1tpk2tpk3 Figure 5-5. The role of Ras2p and PKA in PAQR-dependent repression: BY4742 wild type, ras2 tpk1 tpk2 and tpk3 strains are used in panels A and B. -Galactosidase activities are shown as a percentage of fully induced activity in wild type strain carrying an empty control vector grown in iron-deficient and iron-replete LIM. The error bars represent 1 standard deviation for experiments performed in triplicate. (A) The repression of -galactosidase activities of the p FET3-lacZ reporter by overexpression of Izh1p, Izh3p, and Izh4p were absent in the ras2 tpk1 tpk2 and tpk3 mutant strains. (B) The repression of -galactosidase activities of the pFET3-lacZ reporter by overexpression of Adip oR1, AdipoR2 with adiponectin, PAQR3, and PAQR4 were absent in the ras2 tpk1 tpk2 and tpk3 mutant strains. (C) Th e repression of galactosidase activities of the p FET3-lacZ reporter by overexpression of mPR with progesterone, mPR mPR and PAQR11 were absent in the ras2 tpk1 tpk2 and tpk3 mutant strains.

PAGE 143

143 -Galactosidase Activity % of WT Low Fe3+ Vector Control 0 20 40 60 80 100 120 140 160 Control AdipoR1 AdipoR2 PAQR3 PAQR4 Wild Type ras2tpk1tpk2tpk3B -Galactosidase Activity % of WT Low Fe3+ Vector Control 0 20 40 60 80 100 120 140 160 Control mPR mPR mPR PAQR11 Wild Type ras2tpk1tpk2tpk3C Figure 5-5 continued.

PAGE 144

144 -Galactosidase Activity % of WT Low Fe3+ Vector Control 0 20 40 60 80 100 120 140 160 Control Izh1p Izh3p Izh4p Wild Type pkh1pkh2A Figure 5-6. PDK is essential for human PAQR -dependent repression: BY4742 wild type, pdk1 and pdk2 strains are used in panels A to C. -Galactosidase activities are shown as a percentage of fully induced activ ity in wild type stra in carrying an empty control vector grown in iron-deficient and ir on-replete LIM. The error bars represent 1 standard deviation for experiments performed in triplicate. (A) The repression of -galactosidase activities of the p FET3-lacZ reporter by overexpression of Izh1p, Izh3p, and Izh4p were absent in the pdk1 and pdk2 mutant strains. (B) The repression of -galactosidase activities of the p FET3-lacZ reporter by overexpression of AdipoR1, AdipoR2 with adiponectin, PAQR3, and PAQR4 were absent in the pdk1 and pdk2 mutant strains. (C) The repression of -galactosidase activities of the p FET3-lacZ reporter by overexpression of mPR with progesterone, mPR mPR and PAQR11 were absent in the pkh1 and pkh2 mutant strains.

PAGE 145

145 Wild Type -Galactosidase Activity % of WT Low Fe3+ Vector Control 0 20 40 60 80 100 120 140 160 Control AdipoR1 AdipoR2 PAQR3 PAQR4 pkh1pkh2B -Galactosidase Activity % of WT Low Fe3+ Vector Control 0 20 40 60 80 100 120 140 160 Control mPR mPR mPR PAQR11 Wild Type pkh1pkh2C Figure 5-6. Continued

PAGE 146

146 CHAPTER 6 THE EFFECTS OF ADIPONECTIN AND TUMOR NECROSIS FACTOR ON ADIPOR1 Introduction In the last twenty years, obesity has reached epidemic levels in the United States (Zimmet and Thomas, 2003). Obesity can lead to the problems such as insulin resistance, type 2 diabetes, vascular disorders, and many chronic diseases (Cummings and Schwartz, 2003). Many of these disorders result from altered immune function brought on by an increase of adipose tissue (Channon and Guzik, 2002). Adipose tissue, originally thought as an en ergy storage source, has important roles in endocrine and immune systems (Trujillo and Schere r, 2006). This tissue also plays a role in controlling metabolism by secreti ng a variety of bioactive medi ators known as adipocytokines (Rajala and Scherer, 2003). One of the proteins adiponectin, is found in high levels in adipose tissue, improves insulin sensitivity and decr eases inflammation (Sch erer et al., 1995). Adiponectin levels are decreased in individuals with insulin resi stance and type 2 diabetes, but levels increase after weight loss (Hu and Sp iegelman., 1996). Anothe r protein secreted by adipose tissue is the proinflamm atory tumor necrosis factor(TNF), which is structurally related to adiponectin and has increased levels in obe sity (Klover, et al., 2003). Two adiponectin receptors (AdipoR1 and AdipoR2) exhibit binding to adiponectin (Yamauchi, 2003). Both AdipoR1 and AdipoR2 gene s encode membrane proteins that belong to a large family known as the P rogestin/A dipoQ R eceptors (PAQRs) (Tang, 2005). The PAQRs are comprised of three classes, of which AdipoR1 and AdipoR2 are class I proteins. This family of proteins contains at leas t seven transmembrane domains an d similar predicted topology with the cytoplasmic N-terminus and extra-cytoplas mic C-terminus, except for the class II PAQRs

PAGE 147

147 that contain an eighth transmem brane domain (Tang, 2005). More over, these proteins contain three conserved motifs found on the cyt oplasmic side of the membrane. We previously demonstrated that overexpres sion of AdipoR1 in ye ast has a measurable phenotype (Kupchak et al., 2007). For instan ce, AdipoR1 protein overexpression, using a GAL1 inducible promoter, results in repression in low iron media (LIM) of the gene encoding Fet3p, a component of the high affinity iron-uptake comp lex. In this study, we show that AdipoR1 overexpression with the addition of adiponectin can repress a p FET3-lacZ reporter in low iron media (LIM). Also, TNFcan partially alleviate repre ssion of FET3 caused by AdipoR1 overexpression in iron-deficient LIM. However, when adiponectin and TNFwere added simultaneously, repression was not a lleviated in LIM with galactos e or raffinose. In this study, we also demonstrated--using flow cytome try--that FAM-adiponectin and FITC TNFexhibited binding to AdipoR1 on the surface of yeast spheroplasts. Additionally, we showed that unlabeled adiponectin lowe red binding of FITC TNF; however, FAM-adiponectin binding was only alleviated at high levels of unlabeled TNF. Thus, we report that TNFcan affect and bind to AdipoR1, even though AdipoR 1 prefers binding to adiponectin. Materials and Methods Yeast Strains and Plasmids W ild type (BY4742) and izh1 izh2 izh3 izh4 mutant strains were obtained from Euroscarf and their genotypes are listed in supplementary table 1. The p FET3-lacZ plasmid (Gift of D. Eide) has the promoter region of the FET3 gene fused to the lacZ open reading frame and is URA3 selectable. AdipoR1 and PAQR11 genes were cloned from genomic DNA into pRS316GAL1 ( LEU2 ) via homologous recombination as previously described (Lyons, et al., 2004).

PAGE 148

148 Primers used to clone AdipoR1 and PAQR11 are listed in supplementary table 2. All clones were sequenced to ensure that no errors occurred in the PCR. -Galactosidase Assays -Galactosidase and ferroxidase assays were perfor med as de scribed in chapter 2 except for the addition of the following compounds to LIM. Adiponectin (Biovendor Laboratory Medicine, Inc. #RD172029100), tumor necrosis factor(Chemicon International #GF023) and tumor necrosis factor(Fitzgerald Industries Internationa l #RDI-301BH) were added to the cultures at varied concentrations. Spheroplast Formation Spheroplasts for the izh1 izh2 izh3 izh4 m utant strain were prepared as reported by Holcombe et al (1987), with modifications. A si ngle colony was picked and placed in synthetic dextrose (SD) medium and grown to an OD600 ~ 5.0. The culture was reinoculated into synthetic galactose (SGal) media at an OD600 of 0.2 and grown to OD600 between 1.5 and 2.0. Cells were centrifuged at 4,000 rpm, and washed with 20 mL 0.1 M Tris-Cl, pH 9.4, 10 mM dithiothreitol, 50 mM -mercaptoethanol, and incubated for 30 minutes with gentle shaking at 30C. After centrifugation, the pellet was washed in spheroplasting buffer (1.2 M sorbitol, 10 mM potassium phosphate, 0.5x YPD), centrifuged and resuspende d into spheroplasting buffer with 12,500 units of lyticase at an OD600 of 25. Culture was incubated at 30C with gentle shaking for 45 minutes, and washed twice with 1x PBS. The spheroplas ts were recovered by cen trifuging at 500 rpm for 10 minutes, then taking off the supernantant. The supernatant was diluted with 1x PBS to an OD600 of 0.05. Addition of Fluorophores Mono-5-(and-6)-carboxyfluoresein (FAM)-Adipon ectin (Phoenix Pharm auceticals, Inc. #FG-ADI-01-A) at varied concentrations was ad ded directly to the spheroplast suspension on

PAGE 149

149 ice. The TNFcould not be added directly; therefor e a commercially available kit (Human TNFBiotin Conjugate Kit, R&D Systems #NFT A0) designed to detect binding of TNFto humans was successfully modified to dete ct binding to yeast spheroplasts. 10 L of biotinylated rhTNFat varied concentrations wa s added to the spheroplast su spension and incubated for 60 minutes at 4C. To the spheroplast suspension, 10 L of avadin-fluorescein isothiocyanate (FITC) reagent was added and then incubated for 30 minutes at 4C in the dark. Spheroplasts were washed twice with 1x RDF1 buffer and resu spended in 0.2 mL of 1x RDF1 buffer. For competitive binding assays, nonlabeled adiponectin, TNF, or TNFwas added at varied concentrations to the spheroplast suspension. Flow Cytometry Flow cytometry analysis was run on a FACS Vant age SE Turbosort (BD Biosciences, San Jose, CA); the ins trument utilizes Cell Quest software from BD Biosciences to collect and analyze data. Data was shown as % of Ligand Binding to Spheroplasts and signifies percent of bound ligand to spheroplasts with AdipoR1 or PAQR 11 overexpressed, minus percent of bound ligand to control cells. Percen t of ligand is a measurement of num ber of cells bound to the fluorophore, divided by 50,000 cells counted, multiplied by 100. In order to measure the number of cells bound to fluorophore, unlabed cells were initially measured and visualized in a scattergraph, which has two regions. The unlabeled cells we re placed in region R2. As the fluorophore was added, binding to the specific ligand by the cells was shifted to region R1. R1 then would be divided by total number of cells counted and tu rned into a percent to give ligand bound to control, AdipoR1, or PAQR11. Results The surface location of an adiponectin recepto r was established in previous studies, as localized to the plasma membrane in 3T3L1 adipocytes (Bogan and Lodish., 1999).

PAGE 150

150 Adiponectin has been shown to bind to AdipoR 1 and activate AMP-kinase (AMPK) in mouse skeletal muscle (Yamauchi, 2003; Yamauchi, 2002) In yeast, AdipoR1 overexpression signals through AMPK to repress FET3 expression in galactose contai ning iron-deficient LIM (Kupchak et al., 2007). Therefore, we examined ad iponectins ability to elicit an effect on FET3 We also examined the effect of TNFon FET3 during AdipoR1 overexpr ession, because we found sequence homology performing BLAST searches between adiponectin and TNF. To see if adiponectin and TNFbinds to AdipoR1 in yeast, we made spheroplasts that gave us access to the plasma membrane. Effect of AdipoR1 on FET3 Expression The p FET3-lacZ reporter responds to the amount of ir on added to LIM. It is fully induced during iron-deficiency (1 M) and repressed during iron-replete (1 mM) conditions. The induction during iron-defic iency is repressed by AdipoR1 gene overexpression (Figure 61A). We also demonstrated that repression of p FET3-lacZ by AdipoR1 overexpression can be incrementally diminished by decreasing the gala ctose concentration, whic h is the inducer of PAQR expression (Figure 6-1B). Adiponectin Binds to AdipoR1 on Yeast Spheroplasts pFET3-lacZ expression is restored at low concentrations (0.05%) of galactose. However, activ ity of p FET3-lacZ starts to decline at 0.1 pM adi ponectin and is repressed at 10 pM adiponectin (Figure 6-2A). To ensure that adiponectin binds to AdipoR1 overexpressed on spheroplasts, we performed fl ow cytometry studies using FAM-adiponectin. Binding to AdipoR1 was exhibited at 1 pM, and demonstr ated saturation at 20 pM (Figure 6-2B). TNFbinds to AdipoR1 on Spheroplasts We examined whether TNFexhibited an effect on p FET3-lacZ activity when AdipoR1 is overexpressed in galactose co ntaining LIM. Repression of p FET3-lacZ caused by TNFis

PAGE 151

151 partially relieved at 1 nM a nd exhibited its maximal effect at 50 nM (Figure 6-3A). TNFwas also examined, but showed no respon se (Figure 6-3A). Binding of TNFto AdipoR1 overexpressed on spheroplasts show s binding as low as 1nM and saturates the receptor at 40 nM (Figure 6-3B). Adiponectin, rather than TNF, prefers bin ding to AdipoR1 To examine whether AdipoR1 prefer entially binds adiponectin or TNF, we performed competitive studies where the addition of adiponectin and TNFwere measured their effect using -Galactosidase and flow cytometry binding assays. p FET3-lacZ repression in raffinose containing LIM by AdipoR1 overexpression was repressed by 100 pM adiponectin. This repression was not effected by the additi on of increasing concentrations of TNF(Figure 6-4A). For flow cytometry studies, binding of th e fluorophore, FAM-adiponectin to AdipoR1 on spheroplasts was affected only by the addition of 50 nM of TNF(Figure 6-4B). The FITCTNFshowed less AdipoR1 binding when increasing levels of adiponectin were added (Figure 6-4C). Discussion Adiponectin receptor 1 is expressed in many tissues, but is highly enriched in skeletal muscle. AdipoR1 is a 375 amino acid protein (42.4 kDa) consisting of seven transmembrane domains with an intracellular N-terminus (Y amauchi, 2003). AdipoR1 belongs to a family known as the class I PAQRs (Lyons et al., 2004). The PAQRs ar e conserved from yeast to humans (Klover et al., 2003). The yeast homol ogue, Izh2p, acts similarly to AdipoR1, regulating lipid metabolism through AMPK activation (Y amauchi, 2003) (Karpi chev et al., 2002). Effect of AdipoR1 on Fet3 expres sion In previous studies, we demonstrated th at Izh1p, Izh2p, Izh3p a nd Izh4p activities could be monitored by using -galactosidase assays, and that overexpression of any of the Izh proteins

PAGE 152

152 negatively repress FET3 This effect was not due to a gene ralized growth defect, expression of any random protein, or a global defect in transcription, translation, or -galactosidase activity (Lyons et al., 2004). Since yeast does not encode an adiponectin receptor, we overexpressed the AdipoR1 gene to study the protein. Similarly to Izh2p, AdipoR1 represses FET3 in the absence of its activating ligand adipon ectin (Kupchak et al., 2007) It is possible that AdipoR1 in yeast has an unidentified endogenous liga nd, and this hypothetical molecu le may be present in a high concentration to activate AdipoR 1 when it is overexpressed. It is also possible that AdipoR1, like many signal proteins, has an intrinsic basa l signaling capability that is amplified by overexpression, making the presence of an activati ng ligand unnecessary. In the latter case, overexpression would function equivalently to activa tion. To test this model, we performed the experiment with raffinose, instead of galactose containing LIM. We postulate raffinose decreased the expression of th e AdipoR1 protein, thereby inhi biting activation and limiting the signaling capability to repress FET3 Adiponectin binding to AdipoR1 on spheroplasts Adiponectin is a protein secr eted by adipocytes and found at high concentrations (g/mL) in healthy individuals. Unlike other adipocyt okines, whose levels increase with fat mass, adiponectin concentration decrease s with obesity (Arita et al., 1999). Adi ponectin levels also drop in patients with type 2 diabetes (Hotta et al., 2000), people with re duced vascular function, and people with increased cornary artery disease (Goldstein and Scalia, 2004). Adiponectin structurally is a heteorotrimer (244 amino acids, 24 kDa), consisting of a Nterminal collagen-like sequence and a C-termin al globular region (Fru ebis et al., 2001) and belongs to the complement C1q family (Crouch, 1994; McCormack et al., 1997; Wong et al., 2004).

PAGE 153

153 Unlike Izh2p, AdipoR1 has a commercially avai lable ligand-adiponectin. Therefore, we tested ligand-receptor activation. Increasing concentrations of adiponectin added to AdipoR1 overexpressing cells in raffinose containing LIM caused an incremental repression of FET3 Even though AdipoR1 protein expression was lowered, FET3 repression may occur due to the ligand binding and activating the signaling pathway. We proceeded to test binding of adiponectin to AdipoR1 on spheropl asts. Spheroplasts were used because AdipoR1 was localized to the plasma membrane (Yamauchi, 2003) and because yeast contain a cell wall, which must be removed to allow efficient binding. To examine binding, fluorescently labeled adi poectin was added and exhibited a high level of binding to cells expressing AdipoR1. PAQR11, a class III PAQR homologue, also was investigated due to its sequence relation to AdipoR1 (Klover et al., 200 3) and its supposed inability to bind to adiponectin. As expected, adiponectin only exhib ited binding to AdipoR1 at high concentrations and this could be attributed to nonspecific binding to the spheroplasts. This result indicates that yeast spheroplasts are a viable model to examine human membrane receptor binding. TNFbinding to AdipoR1 on spheroplasts TNFis a 17 kDa soluble protein secreted by adipocytes. Its level is increased during diabetes (Wong et al., 2006). TNFwas exam ined because it showed homology to adiponectin and features common to all C1q proteins (Scherer et al., 1995). Also, other ligands have been shown to bind to AdipoR1. Osmotin, a receptor for Izh2p (Narasimhan et al., 2005), has no sequence homology with adiponectin; however, it show s structural similarity to adiponectin and can bind and activate AdipoR1 (Narasimhan et al., 2005). TNFwas tested to examine an effect on FET3 repression. TNFexhibited an increase of FET3 induction, with an increase in concentration of TNF. We postulate TNFmay act on AMPK via AdipoR1 and cause localization to the nucle us, and repress the negative regulators of FET3 Nrg1p and Nrg2p,

PAGE 154

154 thereby inducing FET3. In the literature, TNFhas been shown to reduce expression of adiponectin from 3T3-L1 adipoc ytes (Fasshauer et al., 2002); thereby AdipoR1 overexpression may be unable to activate signaling and repress FET3 TNFwas also studied because it exhibited structurally similarity to TNF, as a pear/cone shaped trimer (Eck and Sprang, 1989; Eck et al., 1992). However, unlike TNF, TNFwas unable to cause an effect on FET3 This may be a result from TNFonly sharing 30% identity with TNF(Pennica et al., 1984). Since, TNFalleviated FET3 repression caused by AdipoR1, we examined TNFbinding to AdipoR1 on spheroplasts. TNFexhibited binding to AdipoR1, but at a 1,000 times greater concentration than ad iponectin. Even though TNFshowed a lower binding percentage at a greater concentration, th is was expected since TNFis not AdipoR1s native ligand. However, this is a significant result because it showed TNFcan bind to AdipoR1. Adiponectin, rather than TNF, prefers bin ding to AdipoR1 Adiponectin and TNFare two cytokines secreted by ad ipocytes; however, they have a reciprocal relationship in humans (Maeda et al., 2002). Pr evious studies have shown normal concentrations of ad iponectin suppress TNFin macrophages (Ouchi et al., 2000). Therefore, we investigated adiponectin and TNFsimultaneous effect on FET3 when AdipoR1 was overexpressed. Results showed adiponectin at lower concentrations than TNFstill repressed FET3 To verify these results, we also investig ated the effect of an unlabeled ligand on a fluorinated ligands ability to bind to AdipoR1. In both situat ions, AdipoR1 preferentially bound to adiponectin. This effect on FET3 and on binding was not a surprise since adiponectin is AdipoR1s native ligand. Although this study e xhibited binding of adiponectin and TNFto the receptor, we were unable to calculate the bindi ng affinity of these two ligands to the receptor because protein receptor concentration could not be obtained.

PAGE 155

155 Summation Our findings support three m ain conclusions First, AdipoR1 activates a signal transduction pathway that represses FET3 expression, while the addition of TNFalleviates FET3 repression. Second, AdipoR1, overexpressed on the spheroplasts surface, binds adiponectin and TNF. Third, AdipoR1 preferentially binds adiponectin over TNF. Future experiments will focus on two main questions. First, can TNFbind to AdipoR1 in mammalian cells, or does binding to TNFonly occur in yeast? Seco nd, what is the effect of adiponectin and TNFon AMPK activity when AdipoR1 is overexpressed? We predict an increase in AMPK activity with the addition of adiponectin and a decrease in activity when TNFis added. If there is an effect on AMPK by TNF, then we can state that TNFnot only binds to AdipoR1, but can also alleviate FET3 repression through AMPK deactivation.

PAGE 156

156 -Galactosidase Activity % of WT Low Fe3+ Vector Control 0 20 40 60 80 100 120 1 M Fe3+ 1,000 M Fe3+ Control AdipoR1A Figure 6-1. AdipoR1 represses FET3 transcription: BY4742 wild type strain is used in panels A and B. -Galactosidase activities are shown as a percentage of fully induced galactosidase activity in wild type strain carrying an empty control vector grown in iron-deficient LIM. The error bars repres ent 1 standard deviation for experiments performed in triplicate. (A) -Galactosidase activities of the p FET3-lacZ reporter are induced in LIM in the wild type strain carrying the AdipoR1 overexpressor. (B) The AdipoR1 effect on p FET3-lacZ decreases as the percent of galactose in iron-deficient LIM is decreased. Activities for each ga lactose concentration are normalized to percent of activity in a strain carrying an empty control vector grown at the same concentration. 0.0 -Galactosidase Activity % of Low Fe3+Vector Control 0 20 40 60 80 100 120 AdipoR1 0.01 0.03 0.05 0.1 0.30.5 0.75 1.0 1.5 2.0 % of GalactoseB Figure 6-1. Continued.

PAGE 157

157 0.0 -Galactosidase Activity % of Low Fe3+Vector Control 0 20 40 60 80 100 120 Control AdipoR1 0.11.0 10 100 1,000 pM [Adiponectin]A Figure 6-2. Adiponectin binds to AdipoR1 on spheroplasts: For panel A, -galactosidase activities are shown as a percentage of fully induced -galactosidase activity in wild type strain carrying an empt y control vector grown in iron-deficient LIM. For panel B, binding is shown as percentage of FAMadiponectin binding to spheroplasts in the izh1 izh2 izh3 izh4 strain carrying an AdipoR1 or PAQR11 overexpressor, minus the percentage of FAM-adiponec tin binding to spheroplasts in the izh1 izh2 izh3 izh4 strain carrying the empty control vector in SGal For both panels, the error bars represent 1 standard deviati on for experiments performed in triplicate. (A) AdipoR1 overexpression in 1.98% raffinos e/0.02% galactose exhibits no effect on the p FET3-lacZ reporter unless adipon ectin is added, while adiponectin alone has no effect. For cells grown in 1.98% ra ffinose/0.02% galactose, activities are normalized to percent of activity in a wild type strain carrying the empty control vector grown in iron-deficient LIM with the same concentration of raffinose/galactose. (B) FAM-adiponectin in SGal exhibits binding to AdipoR1 overexpressed spheroplasts, while PAQR11 ove rexpressed spheroplasts demonstrates a low percentage of binding to FAM-adiponectin.

PAGE 158

158 0.0 % of FAM-Adiponectin Binding to Spheroplasts 0 20 40 60 80 100 AdipoR1 PAQR11 0.1 1.0 10 20 40 pM [FAM-Adiponectin]B Figure 6-2 continued.

PAGE 159

159 -Galactosidase Activity % of Low Fe3+Vector Control 0 20 40 60 80 100 120 Control with TNFControl with TNFAdipoR1 with TNFAdipoR1 with TNF0.0 0.01 0.11.0 10 100 nM [Ligand] 50A Figure 6-3. TNFbinds to AdipoR1 on spheroplasts: For panel A, -galactosidase activities are shown as a percentage of fully induced -galactosidase activity in wild type strain carrying an empty control vector grown in iron-deficient LIM. For panel B, binding is shown as percentage of FITC TNFbinding to spheroplasts in the izh1 izh2 izh3 izh4 strain carrying an AdipoR1 or PAQR11 overexpressor, minus the percentage of FITC TNFbinding to spheroplasts in the izh1 izh2 izh3 izh4 strain carrying the empty control vector in SGal. For both panels, the error bars represent 1 standard deviation for expe riments performed in triplicate. (A) Galactosidase activities of the p FET3-lacZ reporter are partiall y induced in the wild type strain carrying the Ad ipoR1 overexpressor when TNFis added, but is not induced by the addition of TNF. The empty control vector in the wild type strain was induced when TNFor TNFwas added. (B) FITC TNFin SGal exhibits binding to AdipoR1 overexpressed in spheroplasts, while PAQR11 overexpressed in spheroplasts shows a low percen tage of binding to adiponectin.

PAGE 160

160 0.0 % of FITC TNFBinding to Spheroplasts 0 10 20 30 40 50 60 70 AdipoR1 PAQR11 0.1 1.0 10 40 80BnM [FITC TNF) Figure 6-3. Continued.

PAGE 161

161 -Galactosidase Activity % of Low Fe3+ Vector Control in Raffinose 0 20 40 60 80 100 120 Control AdipoR1p nM [TNF] 0.0 0.1 50 1.0 10 100 pM Adiponectin 100A Figure 6-4. Competition of binding to AdipoR1 between adiponectin and TNF: For panel A, -galactosidase activities are shown as a percentage of fully induced -galactosidase activity in wild type strain carrying an em pty control vector grown in iron-deficient LIM. For panels B and C, binding is show n as percentage of FAM-adiponectin or FITC TNFbinding to spheroplasts in the izh1izh2 izh3 izh4 strain carrying the AdipoR1 overexpressor, minus the percenta ge of FAM-adiponectin or FITC TNFbinding to spheroplasts in the izh1 izh2 izh3 izh4 strain carrying the empty control vector in SGal. For panel A to C, the erro r bars represent 1 st andard deviation for experiments performed in triplicate. (A) AdipoR1 ove rexpressor in 1.98% raffinose/0.02% galactose s hows decreased activity of -galactosidase when adiponectin is added, but the addition of TNFdid not further affect -galactosidase activity. For cells grown 1.98% raffinose/0.02 % galactose, activities are normalized to percent of activity in a wild type strain grown in iron-deficient LIM with the same concentration of raffinose/galactose. (B) FAM-adipon ectin exhibits binding to AdipoR1 overexpressed in spheroplasts, and are effected by the a ddition of unlabeled TNFat high concentrations, whereas the addition of TNFdid not affect FAMadiponectin binding to AdipoR1 overexpre ssed in spheroplasts. (C) FITC TNFexhibits binding to AdipoR1 overexpressed in spheroplasts, but binding is lowered with the addition of unlabeled adiponectin.

PAGE 162

162 % of FAM-Adiponectin Binding to Spheroplasts 0 20 40 60 80 100 AdipoR1 with TNFAdipoR1 with TNFnM [TNFor TNF] 0.0 10 100 50 20 pM FAM-Adiponectin 1.0 0.1 5.0B Figure 6-4. Continued. % of TNFBinding to Spheroplasts 0 10 20 30 40 50 60 AdipoR1 pM [Adiponectin] 0.0 10 30 20 40 40 nM FITC TNFC

PAGE 163

163 APPENDIX A YEAST STRAINS UTILIZED THOUGHOUT THIS STUDY Strain Mutation Source Genotype BY4742 Wild Type Euroscarf MAT ; his3 1; leu2 0; lys2 0; ura2 0 W303-1A Wild Type Dr. Howard Riezman MAT ; his3 1; leu2 0; ura2 0 MCY5326 Wild Type Dr. Marion Carlson MAT ; his3 1; leu2 0; ura2 0 YDR490w pkh1 Euroscarf MAT ; his3 1; leu2 0; lys2 0; ura2 0 pkh1 ::kanMX4 YOL100c pkh2 Euroscarf MAT ; his3 1; leu2 0; lys2 0; ura2 0 pkh2 ::kanMX4 YNL098c ras2 Euroscarf MAT ; his3 1; leu2 0; lys2 0; ura2 0 ras2 ::kanMX4 YJL164c tpk1 Euroscarf MAT ; his3 1; leu2 0; lys2 0; ura2 0 tpk1 ::kanMX4 YPL203w tpk2 Euroscarf MAT ; his3 1; leu2 0; lys2 0; ura2 0 tpk2 ::kanMX4 YKL166c tpk3 Euroscarf MAT ; his3 1; leu2 0; lys2 0; ura2 0 tpk3 ::kanMX4 YMR037c msn2 Euroscarf MAT ; his3 1; leu2 0; lys2 0; ura2 0 msn2 ::kanMX4 YKL062w msn4 Euroscarf MAT ; his3 1; leu2 0; lys2 0; ura2 0 msn4 ::kanMX4 YDR477w snf1 Euroscarf MAT ; his3 1; leu2 0; lys2 0; ura2 0 snf1 ::kanMX4 YGL115w snf4 Euroscarf MAT ; his3 1; leu2 0; lys2 0; ura2 0 snf4 ::kanMX4 YER027c gal83 Euroscarf MAT ; his3 1; leu2 0; lys2 0; ura2 0 gal83 ::kanMX4 YDR422c sip1 Euroscarf MAT ; his3 1; leu2 0; lys2 0; ura2 0 sip1 ::kanMX4 YGL208w sip2 Euroscarf MAT ; his3 1; leu2 0; lys2 0; ura2 0 sip2 ::kanMX4 YKL048c elm1 Euroscarf MAT ; his3 1; leu2 0; lys2 0; ura2 0 elm1 ::kanMX4 YGL179c tos3 Euroscarf MAT ; his3 1; leu2 0; lys2 0; ura2 0 tos3 ::kanMX4 YER129w sak1 Euroscarf MAT ; his3 1; leu2 0; lys2 0; ura2 0 sak1 ::kanMX4 YDR043c nrg1 Euroscarf MAT ; his3 1; leu2 0; lys2 0; ura2 0 nrg1 ::kanMX4 YBR066c nrg2 Euroscarf MAT ; his3 1; leu2 0; lys2 0; ura2 0 nrg2 ::kanMX4

PAGE 164

164 YOR171c lcb4 Euroscarf MAT ; his3 1; leu2 0; lys2 0; ura2 0 lcb4 ::kanMX4 YLR260w lcb5 Euroscarf MAT ; his3 1; leu2 0; lys2 0; ura2 0 lcb5 ::kanMX4 YDR297w sur2 Euroscarf MAT ; his3 1; leu2 0; lys2 0; ura2 0 sur2 ::kanMX4 YMR272c scs7 Euroscarf MAT ; his3 1; leu2 0; lys2 0; ura2 0 scs7 ::kanMX4 YPL057c sur1 Euroscarf MAT ; his3 1; leu2 0; lys2 0; ura2 0 sur1 ::kanMX4 YBR036c csg2 Euroscarf MAT ; his3 1; leu2 0; lys2 0; ura2 0 csg2 ::kanMX4 YDR072c ipt1 Euroscarf MAT ; his3 1; leu2 0; lys2 0; ura2 0 ipt1 ::kanMX4 YCR034w fen1 Euroscarf MAT ; his3 1; leu2 0; lys2 0; ura2 0 fen1 ::kanMX4 YML052w sur7 Euroscarf MAT ; his3 1; leu2 0; lys2 0; ura2 0 sur7 ::kanMX4 MCY5338 msn2 msn4 Dr. Marion Carlson MAT ; his3 1; leu2 0; ura2 0 msn2 ::kanMX6; msn4 ::natMX4 MCY5378 nrg1 nrg2 Dr. Marion Carlson MAT ; his3 1; leu2 0; ura2 0 nrg1 ::hphMX4; nrg2 ::his3 MX6 MCY5385 msn2 msn4 Dr. Marion Carlson MAT ; his3 1; leu2 0; ura2 0 nrg1 nrg2 nrg1 ::hphMX4; nrg2 ::his3MX6

PAGE 165

165 APPENDIX B PRIMERS USED FOR CLONING pRS316GA L1-IZH1 5 5 TAC TTC TTA TTC CTC TAC CGG ATC CCG CTC GAG GTC GAC ATG AGT ATC ACC ACT ACT AG 3 pRS316GAL1-IZH1 3 5 TGA GCG CGC GTA ATA CGA CTC ACT ATA GGG CGA ATT GGA GCT CGT AGT AGT TGT AGG CCC A 3 pRS316GAL1-IZH2 5 5 TAC TTC TTA TTC CTC TAC CGG ATC CCG CTC GAG GTC GAC ATG TCA ACT TTA TTA GAA AGG pRS316GAL1-IZH2 3 5 TGA GCG CGC GTA ATA CGA CTC ACT ATA GGG CGA ATT GGA GCT CCA AAT ATC TAG GAG ACA AT pRS316GAL1-IZH3 5 5 TAC TTC TTA TTC CTC TAC CGG ATC CCG CTC GAG GTC GAC ATG ATG GAC TCG TCA AGC A pRS316GAL1-IZH3 3 5 TGA GCG CGC GTA ATA CGA CTC ACT ATA GGG CGA ATT GGA GCT CAG GAT GTT GTA AGT AAA GG pRS316GAL1-IZH4 5 5 TAC TTC TTA TTC CTC TAC CGG ATC CCG CTC GAG GTC GAC ATG GTT TCA TTG ACT ACA ATA pRS316GAL1-IZH4 3 5 TGA GCG CGC GTA ATA CGA CTC ACT ATA GGG CGA ATT GGA GCT CAC CAG CTG AGT CTA AGA pRS316GAL1AdipoR1 5 5TAC TTC TTA TTC CTC TAC CGG ATC CCG CTC GAG GTC GAC ATG TCT TCC CAC AAA GGA 3 pRS316GAL1AdipoR1 3 5TGA GCG CGC GTA ATA CGA CTC ACT ATA GGG CGA ATT GGA GCT CTC AGA GAA GGG TGT CAT CA pYES260GAL1AdipoR2 5 5GGT GGT GGC GAC CAT CAC GAG AAT CTT TAT TTT CAG GGC GCC ATG GGC ATG TCC CCT CTC TT

PAGE 166

166 pYES260GAL1AdipoR2 3 5ATA TCT GCA GAA TTC CAG CAC ACT GGC GGC CGT TAC TAG ATC TTC ACA GTG CAT CCT CTT CAC TGC pRS316GAL1PAQR3 5 5 TAC TTC TTA TTC CTC TAC CGG ATC CCG CTC GAG GTC GAC ATG CAT CAG AAG CTG CTG CTG AA pRS316GAL1PAQR3 3 5 TGA GCG CGC GTA ATA CGA CTC ACT ATA GGG CGA ATT GGA GCT CTC ACA AAT GTG AAA CAT AGT C 3 pRS316GAL1PAQR4 5 5TAC TTC TTA TTC CTC TAC CGG ATC CCG CTC GAG GTC GAC ATG GCG TTC CTG GCC GG pRS316GAL1PAQR4 3 5TGA GCG CGC GTA ATA CGA CTC ACT ATA GGG CGA ATT GGA GCT CTC AGT CCC GGG GAC AGG pRS316GAL1mPR 5 5 TAC TTC TTA TTC CTC TAC CGG ATC CCG CTC GAG GTC GAC ATG CTG AGC CTG AAG CT pRS316GAL1mPR 3 5TGA GCG CGC GTA ATA CGA CTC ACT ATA GGG CGA ATT GGA GCT CTG TTG ACA TCT GGC ATG A pRS316GAL1mPR 5 5TAC TTC TTA TTC CTC TAC CGG ATC CCG CTC GAG GTC GAC ATG GCC ATG GCC CAG AAA 3 pRS316GAL1mPR 3 5TGA GCG CGC GTA ATA CGA CTC ACT ATA GGG CGA ATT GGA GCT CTC ACT TGG TCT TCT GAT CAA pRS316GAL1mPR 5 5TAC TTC TTA TTC CTC TAC CGG ATC CCG CTC GAG GTC GAC ATG ACG ACC GCC ATC TTG 3 pRS316GAL1mPR 3 5TGA GCG CGC GTA ATA CGA CTC ACT ATA GGG CGA ATT GGA GCT CTC AGG AAT CTT TCT TGG TCA pRS316GAL1PAQR11 5 5 TAC TTC TTA TTC CTC TAC CGG ATC CCG CTC GAG GTC GAC ATG CGG TTC AAG AAT CGA TT 3

PAGE 167

167 pRS316GAL1PAQR11 3 5 TGA GCG CGC GTA ATA CGA CTC ACT ATA GGG CGA ATT GGA GCT CTC ATA AAT GCC GCA TAA AGT 3

PAGE 168

168 APPENDIX C MEDIA ABREVIATIONS AND COMPOSITION YPD = Yea st Peptone Dextrose 10.0 g/L yeast extract [BD 2010-09-30] 20.0 g/L peptone [BD 2009-08-17] SD = Synthetic Dextrose 1.7g/L yeast nitrogen base without amino aci ds or ammonium sulfate [Difco 2011-05-31] 5.0g/L ammonium sulfate [Fisher A702-3] 2% w/v glucose [Acros Organics 207-757-8] 0.01% appropriate amino acids [Sigma-Aldrich] SGal = Synthetic Galactose 1.7g/L yeast nitrogen base without amino aci ds or ammonium sulfate [Difco 2011-05-31] 5.0g/L ammonium sulfate [Fisher A702-3] 2% w/v galactose [Acros Organics 200-416-4] 0.01% appropriate amino acids [Sigma-Aldrich] LZM = Low Zinc Media (1 L) 1.7g/L yeast nitrogen base without amino aci ds or ammonium sulfate [Difco 2011-05-31] 5.0g/L ammonium sulfate [Fisher A702-3] 2% w/v galactose [Acros Organics 200-416-4] 0.01% appropriate amino acids [Sigma-Aldrich] 20 mL 1 M sodium citrate, pH 4.2 0.2 mL 10 mM manganese chloride 0.2 mL 10 mM iron chloride 2 mL 500 mM EDTA, pH 8.0 Zinc deficiency, or zinc repletion was made by adding 1 M or 1 mM zinc sulfate Nano pure water was added to 1 L LIM = Low Iron Media (1 L) 1.7g/L yeast nitrogen base without amino aci ds or ammonium sulfate [Difco 2011-05-31] 5.0g/L ammonium sulfate [Fisher A702-3] 2% w/v galactose [Acros Organics 200-416-4] 0.01% appropriate amino acids [Sigma-Aldrich] 20 mL 1 M sodium citrate, pH 4.2 0.2 mL 10 mM manganese chloride 0.2 mL 10 mM zinc sulate 2 mL 500 mM EDTA, pH 8.0 Iron deficiency, or iron re pletion was made by adding 1 M or 1 mM iron chloride Nano pure water was added to 1 L

PAGE 169

169 LIST OF REFERENCES Alkhalaf M, El-Mowafy A, and Karam S (2002) Growth inhibition of M CF-7 human breast cancer cells by progesterone is associated with cell different iation and phosphorylation of Akt protein. Eur J Cancer Prev 11: 481-488. Arita Y, Kihara S, Ouchi N, Takahashi M, M aeda K, Miyagawa J, Hotta K, Shimomura I, Nakamura T, Miyaoka K, Kuriyama H, Nishida M, Yamashita S, Okubo K, Matsbara K, Muraguchi M, Ohmoto Y, Funahashi T, and Ma tsuzawa Y (1999) Paradoxical decrease of an adipose-specific protein, adiponectin, in obesity. Biochem Biophys Res Commun 257: 79-83. Askwith C, Eide DJ, Van Ho A, Bernard PS, Li L, Davis-Kaplan S, Sipe DM, and Kaplan J (1994) The FET3 gene of S. cerevisiae encodes a multicopper oxidase required for ferrous iron uptake. Cell 76: 403-410. Askwith C and Kaplan J (1997) An oxidase-p ermease-based iron transport system in Schizosaccharomyces pombe and its expression in Saccharomyces cerevisiae. J Biol Chem 272: 401-405. Babcock M, de Silva D, Oaks R, Davis-Kaplan S, Jiralerspong S, Montermini L, Pandolofo M, and Kaplan J (1997) Regulation of mito chondrial iron accumulation by Yfh1p, a putative homolog of frataxin. Science 276: 1709-1712. Baida GE and Kuzmin NP (1996) Mechanism of action of hemolysin III from Bacillus cereus. Biochim Biophys Acta 1284: 122-124. Beeler T, Bacikova D, Gable K, Hopkins L, Johnson C, Slife H, and Dunn T (1998) The Saccharomyces cerevisiae TSC10/YBR265w gene encoding 3-ketosphinganine reductase is identified in a screen for temperat ure-sensitive suppressors of the Ca2+-sensitive csg2 mutant. J Biol Chem 273: 30688-30694. Berkey CD, Vyas VK, and Carlson M (2004) Nr g1 and nrg2 transcriptio nal repressors are differentially regulated in response to carbon source. Eukaryot Cell 3 : 311-317. Betz C, Zajonc D, Moll M, and Schweizer E (2002) ISC1 -encoded inositol phosphosphingolipid phospholipase C is involved in Na+/Li+ halotolerance of Saccharomyces cerevisiae Eur J Biochem 269: 4033-4039. Beyersmann D and Harase H (2001) Function of zinc signaling, pr oliferation and differentiation of mammalian cells. Biometals 14: 331-341. Bielwaska A, Greenberg MS, Perry D, Jaya dev S, Shayman JA, McKay C, and Hannun YA (1996) (1S, 2R)-D-erythro-2-(N-myristoylamino )-1-phenyl-1-propanol as an inhibitor of ceramidase. J Biol Chem 271: 12646-12654. Bogan JS and Lodish H.F (1999) Two Compartmen ts for Insulin-stimulated Exocytosis in 3T3L1 Adipocytes Defined by Endogenous ACRP30 and GLUT4. J Cell Biol 146: 609-620.

PAGE 170

170 Boonyaratanakornkit V and Edwards DP (2007) Receptor mechanisms mediating non-genomic actions of sex steroids. Semin Reprod Med 25: 139-153. Boy-Marcotte E, Perrot M, Bussereau F, Boucherie H, and Jacquet M (1998) Msn2p and Msn4p control a large number of genes induced at the diauxic transition which are repressed by cyclic AMP in Saccharomyces cerevisiae. J Bacteriol 180 : 1044-1052. Brauer AU, Nitsch R, and Savaskan NE ( 2004) Identification of macrophage/microglia activation factor (MAF) associated with late endosomes/lysosomes in microglial cells. FEBS Lett 563: 41-48. Brown KH, Peerson JM, and Allen LH (1998) E ffects of zinc supplementation on childrens growth: a meta-analysis of intervention trials. Bibl Nutr Dieta 54: 76-83. Buede R, Rinker-Schaffer C, Pinto WJ, Le ster RL, and Dicks on RC (1991) Cloning and characterization of LCB1, a Saccharomyces cerevisiae gene required for biosynthesis of the long-chain base component of sphingolipids. J Bacteriol 173 : 4325-4332. Calenza JL and Carlson M (1986) A yeast gene that is essent ial for release from glucose repression encodes a protein kinase. Science 233 : 1175-1180. Calenza JL and Carlson M (1989) Mutational analysis of the Saccharomyces cerevisiae SNF1 protein kinase and evidence for functiona l interaction with the SNF4 protein. Mol Cell Biol 9: 5034-5044. Casamayor A, Torrance PD, Kobyashi T, Thor ner J, and Alessi DR (1999) Functional counterparts of mammalian protein kinases PDK1 and SGK in budding yeast. Curr Biol 9: 186197. Chandran M, Phillips SA, and Ciaraldi T, and Henry RR (2003) Adiponectin: more than just another fat cell hormone? Diab Care 26:2442-2450. Channon KM and Guzik TJ (2002) Mechanisms of superoxide production in human blood vessels: relationship to endothelial dysfunction, clini cal and genetic risk factors. J Physiol Pharmacol 53: 515-524. Chen YC, Chang MC, Chuang YC, and Jeang CL (2004) Characterizati on and virulence of hemolysin III from Vibrio vulnificis Curr Microbiol 49: 175-179. Cho RJ, Campbell MJ, Winzeler EA, Steinmetz L, Conway A, Wodicka L, Wolfsberg TG, Gabrielian AE, Landsman D, Lockhart DJ, and Davis RW (1998) A genome-wide transcriptional analysis of the mitotic cell cycle. Mol Cell 2 : 65-73. Correia JN, Conner SJ, and Kirkm an-Brown JC (2007) Non-genomic steroid actions in human spermatozoa. Persistent tickling from a laden environment. Semin Reprod Med 25: 208-219.

PAGE 171

171 Conneely OM, Mulac-Jericevic B, DeMayo F, Lydon JP, and OMalley BW (2002) Reproductive functions of progesterone receptors. Recent Prog Horm Res 57: 339-355. Corton JM, Gillespie JG, Hawley SA, and Hard ie DG (1995) 5-aminoimidazole-4-carboxamide ribonucleoside. A specific method activating AMPactivated protein kina se in intact cells? Eur J Biochem 229: 558-565. Costello LC, Liu Y, Franklin RB, and Kenne dy MC (1997) Zinc inhibition of mitochondrial aconitase and its importance in citrate metabolism of prostate epithelial cells. J Biol Chem 272: 28875-28881. Crouch E, Persson A, Chang D, and Heuser J (1994) Molecular st ructure of pulmonary surfactant. J Biol Chem 269: 17311-17319. Cummings DE and Schwartz MW (2003) Genetics and pathophysiology of human obesity. Annu Rev Med 54: 453-471. Cuvillier O (2002) Sphingosine in apoptosis signaling. Biochim Biophys Acta 1585: 153-162. Dalman PR, Siimes MA, and Stekal A (1980) Iron deficiency in infancy and childhood. Am J Clin Nutr 33 : 86-118. Dancis A, Roman DG, Anderson GJ, Hinnebusch AG, and Klausner RD (1992) Ferric reductase of Saccharomyces cerevisiae: molecular characterization, role in iron uptake, and transcriptional control by iron. Proc Natl Acad Sci USA 89: 3869-3873. De Freitas J, Wintz H, Kim JH, Poynton H, Fox T, and Vulpe C (2003) Yeast, a model organism for iron a copper metabolism studies. Biometals 16 : 185-197. Delaigle AM, Jonas JC, Bauche IB, Cornu O, a nd Brichard SM (2004) I nduction of adiponectin in skeletal muscle by inflammatory cytokines: in vivo and in vitro studies. Endocrinology 145: 5589-5597. De Silva DM, Askwith CC, Eide DJ, and Kaplan J (1995) The FET3 gene product required for high affinity iron transport in y east is a cell surf ace ferroxidase. J Biol Chem 270: 1098-1101. De Vos AM, Hatada M, van der Wel H, Kra bbendam H, Peerdeman AF, and Kim SH (1985) Three-dimensional structure of thauma tin I, an intensity sweet protein. Proc Natl Acad Sci USA 82: 1406-1409. Dickson RC, Nagiec EE, W ells GB, Nagiec MM, and Lester RL (1997) Synthesis of mannose(inositol-P)2-ceramide, th e major sphingolipid in Saccharomyces cerevisiae, requires the IPT1 ( YDR072c ) gene. J Biol Chem 272: 29620-29625. Dickson RC and Lester RL (2002) Sphingolipid functions in Saccharomyces cerevisiae. Biochim Biophys Acta 1583: 13-25.

PAGE 172

172 Dunn TM, Haak D, Monoghan E, and Beeler TJ (1998) Synthesis of monohydroxylated inositolphosphorylceramide (IPC-C) in Saccharomyces cerevisiae requires Scs7p, a protein with both a cytochrome b5-like domain and a hydrolase/desaturase domain. Yeast 14: 311-321. Eck MJ and Sprang SR (1989) The st ructure of tumor necrosis factor-alpha at 2.6 A resolution. Implications for receptor binding. J Biol Chem 264: 17595-17605. Eck M.J, Ultsch M, Rinderknecht E, de Vos AM and Sprang SR (1992) Th e structure of human lymphotoxin (tumor necrosis f actor-beta) at 1.9-A resolution. J Biol Chem 267: 2119-2122. Edwards DP (2005) Regulation of signal transduction pathways by strogen and progesterone. Annu Rev Physiol 67 : 335-376. Eide, DJ (1998) The molecular bi ology of metal ion transport in Saccharomyces cerevisiae Annu Rev Nutr 18: 441-469. Eide, DJ (2001) Functional genomics and metal metabolism. Genome Biol 2: 10-28. Elbing K, McCartney RR, and Schmidt MC (2006) Purificationn and charterization of the three Snf1-activating kinase of Saccharomyces cerevisiae Biochem J 393: 797-805. Ella KM, Qi C, Dolan JW, Thompson RP, a nd Meier KE (1997) Characterization of a sphingomyelinase activity in Saccharomyces cerevisiae. Arch Biochem Biophys 340: 101-110. Fasshauer M, Klein J, Neumann S, Eszlinger M, and Paschke R (2002) Hormonal regulation of adiponectin gene expression in 3T3-L1 adipocytes. Biochem Biophys Res Commun 290 :10841089. Fruebis J, Tsao TS, Javorschi S, Ebbets-Reed D, Erickson MR, Yen FT, Bihain BE, and Lodish HF (2001) Proteolytic cleavag e product of 30-kDa adipocyte complement-related protein increases fatty acid oxidation in musc le and causes weight loss in mice. Proc Natl Acad Sci USA 98: 2005-2010. Futerman AH and Riezman H (2005) The in s and outs of sphingolipid synthesis. Trends Cell Biol 15: 312-318. Gable K, Han G, Monaghan E, Bacikova D, Na tarajan M, Williams R, and Dunn TM (2002) Mutations in the yeast LCB1 and LCB2 genes, including those corr esponding to the hereditary sensory neuropathy type I muta tions, dominantly inactivate se rine palmitoyltransferase. J Biol Chem 277: 10194-10200. Gaigg B, Neergaard TB, Schneiter R, Hansen JK Faergem an NJ, Jensen NA, Anderson JR, Friis J, Schroder HD, and Knudsen J (2001) Depletio n of acyl-coenzyme A-binding protein affects sphingolipid synthesis and causes vesicl e accumulation and membrane defects in Saccharomyces cerevisiae. Mol Biol Cell 12: 1147-1160.

PAGE 173

173 Georgatsou E and Alexandraki D (1994) Two dist inctly regulated genes are required for ferric reduction, the first step of iron uptake in Saccharomyces cerevisiae. Mol Cell Biol 14: 30653073. Gietz RD and Woods RA (2002) Transformati on of yeast by lithium acetate/single-stranded carrier DNA/polyethylene glycol method. Methods Enzymol 350: 87-96. Gitan RS, Shababi M, Kramer M, and Eide DJ (2 003) A cytosolic domain of the yeast Zrt1 zinc transporter is required for its post-translational inactivation in response to zinc and cadmium. J Biol Chem 278: 39558-39564. Gaither LA and Eide DJ (2001) The human ZIP1 transporter me diates zinc uptake in human K562 erythroleukemia cells. J Biol Chem 276: 22258-22264. Goffeau A, Barrell BG, Bussey H, Davis RW, Du jon B, Feldmann H, Galibert F, Hoheisel JD, Jacq C, Johnston M, Louis EJ, Mewes HW, Murakami Y, Philippsen P, Tettelin H, and Oliver SG (1996) Life with 6000 genes. Science 274: 563-567. Goktas S, Yilmaz MI, Caglar K, Sonmez A, K ilic S, and Bedir S (2005) Prostate cancer and adiponectin. Urology 65: 1168-1172. Goldstein BJ and Scalia R (2004) Adiponectin. A novel adipokine linking adipocytes and vascular function. J Clin Endocrinol Metab 89 :2563-2568. Guillas I, Kirchman PA, Chuard R, Pfefferli M, Jiang JC, Jazwinski SM, and Conzelmann A (2001) C26-CoA-dependent ceramide synthesis of Saccharomyces cerevisiae is operated by Lag1p and Lac1p. EMBO J 20: 2655-2665. Haak D, Gable K, Beeler T, and Dunn T (1997) Hydroxylation of Saccharomyces cerevisiae requires Sur2p and Scs7p. J Biol Chem 272: 29704-29710. Han G, Gable K, Kohlwein SD, Beaudoin F, Napier JA, and Dunn TM (2002) The Saccharomyces cerevisiae YBR159w gene encodes the 3-ketoreductase of the microsomal fatty acid elongase. J Biol Chem 277: 35440-35449. Hanada K (2003) Serine palmitoyltransferase, a key enzyme of sphingolipid metabolism. Biochim Biophys Acta 1632: 16-30. Hannun YA (1996) Functions of ceramide in c oordinating cellular response to stress. Science 274: 1855-1859. Hannun YA (1997) Sphingolipid second messengers: tumor suppressor lipids. Adv Exp Med Biol 400: 305-312. Hannun YA and Obeid LM (1997) C eramide and the eukaryotic stress response. Biochem Soc Trends 25: 1171-1175.

PAGE 174

174 Hashida-Okado T, Ogawa A, Endo M, Yasumoto R, Takesako K, and Kato I (1996) AUR1, a novel gene conferring aureobasidin resistance on Saccharomyces cerevisiae: a study of defective morphologies in Aur1p-depleted cells. Mol Gen Genet 251: 236-244. Haurie V, Boucherie H, and Sagliocco F (2003) The Snf1 protein kinase controls the induction of genes of the iron uptake pathway at the diauxic shift in Saccharomyces cerevisiae. J Biol Chem 278: 45391-45396. Hedbacker K, Hong SP, and Carl son M (2004) Pak1 protein kina se regulates activation and nuclear localization of the Snf1-Gal83 protein kinase. Mol Cell Biol 24 : 8255-8263. Holcomb C, Etcheverry T, and Schekman R ( 1987) Isolation of secr etory vesicles from Saccharommyces cerevisiae Anal Biochem 166: 13028-13036. Ho Y, Gruhler A, Heilbut A, Bader GD, Moore L, Adams SL, Millar A, Taylor P, Bennett K, Boutilier K, Yang L, Wolting C, Donaldson I, Sc handorff S, Shewnarane J, Vo M, Taggart J, Goudreault M, Muskat B, Alfarano C, Dewar D, Lin Z, Michalickova K, Willems AR, Sassi H, Nielsen PA, Rasmussem KJ, Anderesen JR, Johansen LE, Hansen LH, Jespersen H, Podtelejnikov A, Nielsen E, Crawford J, Poul sen V, Soresen BD, Matthiesen J, Hendrickson RC, Gleeson F, Pawson T, Moran MF, Dorocher D, Mann M, Hogue CW Figeys D, and Tyers M (2002) Systematic identification of protein complexes in Saccharomyces cerevisiae by mass spectrometry. Nature 415: 180-183. Hotta K, Funahashi T, Arita Y, Takahashi M, Matsuda M, Okamoto Y, Iwahashi H, Kuriyama H, Ouchi N, Maeda K, Nishida M, Kihara S, Sa kai N, Nakajima T, Hasegawa K, Muraguchi M, Ohmoto Y, Nakamura T, Yamashita S, Ha nafusa T, and Matsuzawa Y (2000) Plasma concentrations of a novel, adipose-specific protein, adiponectin, in type 2 diabetic patients. Arterioscler Thromb Vasc Biol 20: 1595-1599. Hu E and Spiegelman BM (1996) AdipoQ is a novel adipose-specific gene dysregulated in obesity. J Biol Chem 271: 10697-10703. Inagaki M, Schmelze T, Yamaguchi K, Irie K, Hall MN, and Matsumoto K (1999) PDK1 homologs activate the Pkc1-mitogen-activat ed protein kinase pathway in yeast. Mol Cell Biol 19: 8344-8352. Ishikawa M, Kitayama J, Kazama S, Hiramats u T, Hatano K, and Nagawa H (2005) Plasma adiponectin and gastric cancer. Clin Cancer Res 11: 466-472. Jiang Y, Davis C, and Broach JR (1998) Effi cient transition to growth on fermentable carbon source in Saccharomyces cerevisiae requi res signaling through the Ras pathway. EMBO J 17: 6942-6951. Kadowaki T and Yamauchi T (2005) Adiponectin and adiponectin receptors. Endocr Rev 26: 439-451.

PAGE 175

175 Karpichev IV, Cornivelli L, and Small GM (2002) Multiple regulatory roles of a novel Saccharomyces cerevisiae protein, encoded by YOL002C, in lipid and phosphate metabolism. J Biol Chem 277: 12152-12162. Kim DH, Sarbassov DD, Ali SM, King JE, Latek RR, Erdjument-Bromage H, Tempst P, and Sabatini DM (2002) mTOR interacts with raptor to form a nutrient-sensitive complex that signals to the cell growth machinery. Cell 110 : 163-175. Kim H, Melen K, and von Heijne GE (2003) Topology models for 37 Saccharomyces cerevisiae membrane proteins based on C-terminal reporter fusions and predictions. J Biol Chem 278: 10208-10213. Kim H, Melen K, Osterberg M, and von He ijne GE (2006) A global topology map of the Saccharomyces cerevisiae membrane proteome. Proc Natl Acad Sci USA 103: 11142-11147. Kline LW and Karpinski E (2005) Progesterone inhibits gallbladder motility through multiple signaling pathways. Steroids 70: 673-679. Klover PJ, Zimmers TA, Koniaris LG, and Moone y RA (2003) Chronic exposure to interleukin6 causes hepatic insulin resistance in mice. Diabetes 11: 2784-2789. Koch J, Gartner S, Li CM, Quintern LE, Bernardo K, Levran O, Schnabel D, Desnick RJ, Schuchman, and Sandhoff K (1996) Molecular cl oning and characterization of a full-length complementary DNA encoding human acid ceramidase Identification of the first molecular lesion causing Farber disease. J Biol Chem 271: 33110-33115. Kohlwein SD, Eder S, Oh CS, Martin CE, Ga ble K, Bacikova D, and Dunn T (2001) Tsc10p is required for fatty acid elongation and localizes to a novel structure at the nuclear-vacuolar interface in Saccharomyces cerevisiae. Mol Cell Biol 21: 109-125. Koiwa H, Kato H, Nakatsu T, Oda J, Yamada Y, and Sato F (1997) Purification and characterization of tobacco pa thogenesis-related protein PR5d, an antifungal thaumatin-like protein. Plant Cell Physiol 38: 783-791. Kruglyak S and Tang H (2000) Regula tion of adjacent yeast genes. Trends Genet 16: 109-111. Kuchin S, Vyas VK, and Carlson M (2002) Snf1 protein kinase and the repressors Nrg1 and Nrg2 regulate FLO11, haploid invasive growth, and diploid pseudohyphal differentiation. Mol Cell Biol 22 : 3994-4000. Kumar A and Snyder M (2001) Genome-wid e transposon mutagenesis in yeast. Curr Protoc Mol Biol 13: 13-16. Kupchak BR, Garitaonandia I, Villa NY, Mulle n MB, W eaver MG, Regalla LM, Kendall EA, and Lyons T.J (2007) Probing the mechanism of FET3 repression by Izh2p overexpression. Biochim Biophys Acta 1773: 1124-1132.

PAGE 176

176 Kurth-Kraczek EJ, Hirshman MF, Goodyear LJ, and Winder WW (1999) 5AMP-activated protein kinase activation causes GLUT4 translocation in skeletal muscle. Diabetes 48 : 16671671. Lesage P, Yang X, and Carlson M (1994) Analysis of the SIP3 protein id entified in atwo-hybrid screen for interaction with the SNF1 protein kinase. Nucleic Acids Res 22 : 597-603. Lisman Q, Pomorski T, Vogelza ngs C, Urli-Stam D, de Cocq van Delwijnen W, and Holthius JC (2004) Protein sorting in the late golgi of Saccharomyces cerevisiae does not require mannosylated sphingolipids. J Biol Chem 279: 1020-1029. Liu K, Zhang X, Sumanasekera C, Lester RL and Dickson RC (2005) Signalling functions for sphingolipid long-chain bases in Saccharomyces cerevisiae Biochem Soc Trans 33: 1170-1173. Lyon CJ, Law RE, and Hsueh WA (2003) Mi nireview: adiposity, inflammation, and atherogenesis. Endocrinology 144: 2195-2200. Lyons, TJ, Gasch AP, Gaither LA, Botstein D, Brown PO, and Eide DJ (2000) Genome-wide characterization of the Zap1p zinc-responsive regulon in yeast. Proc Natl Acad Sci USA 97: 7957-7962. Lyons TJ, Villa NY, Regalla LM, Kupcha k,BR, Vagstad A, and Eide DJ (2004) Metalloregulation of yeast membrane steroid receptor homologs. Proc Natl Acad Sci USA 101: 5506-5511. Maeda N, Shimonura I, Kishida K, Nishizawa H, Matsuda M, Nageretani H, Furuyama N, Kondo H, Takahashi M, Arita Y, Komuro R, Ouchi N, Kihara S, Tochino Y, Okutomi K, Horie M, Takeda S, Aoyama T, Funahashi T, and Matsuzawa Y (2002) Diet-induced insulin resistance in mice lacking adiponectin/ACRP30. Nat Med 8: 731-737. Mandon EC, Ehses I, Rother J, van Echten G, and Sandhoff K (1992) Su bcellular localization and membrane topology of serine palmitoyltransf erase, 3-dehydrosphingosine reductase, and sphinganine N-acyltransferase in mouse liver. J Biol Chem 267: 11144-11148. Mao C, Wadleigh M, Jenkins GM, Hannun YA, and Obeid LM (1997) Identification and characterization of Saccharomyces cerevisiae dihydrosphingosine-1-phosphate phosphatase. J Biol Chem 272: 28690-28694. Mao C, Xu R, Bielawska ZM, and Obeid LM (2000) Cloning and ch aracterization of a Saccharomyces cerevisiae alkaline ceramidase with specificity for dihydroceramide. J Biol Chem 275: 31369-31378. Mao X, Kikani CK, Riojas RA, Langlais P, W ang L, Ramos FJ, Fang Q, Christ-Roberts CY, Hong JY, Kim RY, Liu F, and Dong LQ (2006) APPL1 binds to adiponectin receptors and mediates adiponectin signalling and function. Nat Cell Biol 8 : 516-523.

PAGE 177

177 Mantzoros C, Petridou E, Dessypris N, Chavelas C, Dalamaga M, Alexe DM, Papadiamantis Y, Markopoulos C, Spanos E, Chrousos G, a nd Trichopoulos D (2004) Adiponectin and breast cancer risk. J Clin Endocrinol Metab 89: 1102-1107. Martinez-Pastor MT, Marchler G, Schuller C, Marchler-Bauer A, Ruis H, and Estruch F (1996) The Saccharomyces cerevisiae zinc finger proteins Msn2p and Msn4p are required for transcriptional induction through the stress response element (STRE). EMBO J 15: 2227-2235. Martzen MR, McCraith SM, Spinelli SL, Torres FM, Fields S, Grayhack EJ, and and Phizicky EM (1999) A biochemical genomics approach fo r identifying genes by the activity of their products. Science 286 : 1153-1155. McCormack FX, Pattanajitvilia S, Stewart J, Po ssmayer F, Inchley K, and Voelker DR (1997) The Cys6 intermolecular disulfide bond and the collage n-like region of rat SP-A play critical roles in interactions with alveolar type II cells and surfactant lipids. J Biol Chem 272: 2797127979. Merrill AH Jr, Nimkar S, Menaldino D, Hannun YA, Loomis C, Bell RM, Tyagi SR, Lambeth JD, Stevens VL, and Hunter R ( 1989) Structural requirements for long-chain (sphingolipid) base inhibition of protein kinase C in vitro and for the cellular effects of these compounds. Biochemistry 28: 3138-3145. Merrill GF, Kurth EJ, Hardie DG, and Winde r WW (1997) AICA riboside increases AMPactivated protein kinase, fatty acid oxidat ion, and glucose uptake in rat muscle. Am J Physiol 273: 1107-1112. Miyake Y, Kozutsumi Y, Nakamura S, Fujita T, and Kawasaki T (1995) Serine palmitoyltransferase is the primary target of a sphingosine-like immunosuppresent, ISP1/myriocin. Biochem Biophys Res Commun 211: 396-403. Mora A, Lipina C, Tronche F, Sutherland C, and Alessi DR (2005) Deficiency of PDK1 in liver results in glucose intolerance, impairment of insulin-regulated gene expression and liver failure. Biochem J 385 : 639-648. Mulac-Jericevic B, Mullinax RA, DeMayo FJ, Lydon JP, and Conneeley OM (2000) Subgroup of reproductive functions of progesterone mediated by progesterone-B isoform. Science 289: 1751-1754. Mulac-Jericevic B and Conneel ey OM (2004) Reproductive ti ssue selective actions of progesterone receptors. Reproduction 128: 139-146. Murad AM, dEnfert C, Gaillardin C, Tournu H, Tekaia F, Talibi D, Marechal D, Marchais V, Cottin J, and Brown AJ (2001) Transcript profiling in Candida albicans reveals new cellular functions for the transcriptional re pressors CaTup1, CaMig and CaNrg1. Mol Microbiol 42: 981993.

PAGE 178

178 Nagiec MM, Nagiec EE, Baltisberger JA, Wells GB, Lester RL and Dickson RC (1997) Sphingolipid biosynthesis as a target for antifungal drugs. Co mplementation of the inositol phosphorylceramide synthase defe ct in mutant strain of Saccharomyces cerevisiae by the AUR1 gene. J Biol Chem 272: 9809-9817. Nagiec MM, Skrzypek M, Nagiec EE, Le ster RL, and Dickson RC (1998) The LCB4 ( YOR171c ) and LCB5 ( YLR260w ) genes of Saccharomyces encode sphingoid long chain base kinases. J Biol Chem 273: 19437-19442. Narasimhan ML, Damsz MA, Coca MA, Ibeas JI, Yun DJ, PArdo JM, Hasagawa PM, and Bressen RA (2001) A plant defense respons e effector induces microbial apoptosis. Mol Cell 8: 921-930. Narasimhan ML, Coca MA, Jin J, Yamauchi T, Ito Y, Kadowaki T, Kim KK, Pardo JM, Damsz B, Hasegawa PM, Yun DJ, and Bressan RA (2005) Osmotin is a homolog of mammalian adiponectin and controls apopt osis in yeast through a homol og of mammalian adiponectin receptor. Mol Cell 17 : 171-180. Ogretmen B and Hannun YA (2004) Biologically activ e sphingolipids in cancer pathogenesis and treatment. Nat Rev Cancer 4: 604-616. Oh CS, Toke DA, Mandala S, and Martin CE (1997) ELO2 and ELO3 homologues of the Saccharomyces cerevisiae ELO1 gene, function in fatty acid elongation and are required for sphingolipid formation. J Biol Chem 272: 17376-17384. Oliver SG (1996) From DNA seque nce to biological function. Nature 379: 597-600. Ouchi N, Kihara S, Arita Y, Maeda K, Kurijama H, Okamato Y, Hotta K, Nishida M, Takahashi M, Nakamura T, Yamashita S, Funahashi T, and Matsuzawa Y (1999) Novel modulator for endothelial adhesion molecu les: adipocyte-derived plama protein adiponectin. Circulation 100: 2473-2476. Ouchi N, Kihara S, Arita Y, Okamoto Y, Maeda K, Kurijama H, Hotta K, Nishida M, Takahashi M, Muraguchi M, Ohmoto Y, Nakamura T, Yama shita S, Funahashi T, and Matsuzawa Y (2000) Adiponectin, an adipocyte-derived plasma protein, inhibits endothelial NF-kappaB signaling through a camp-dependent pathway. Circulation 102: 1296-1301. Pajvani UB, Hawkins M, Conbs TP, Rajala MW, Doebber T, Berger JP, Wagner JA, Wu M, Knopps A, Xiang AH, Utzschneid er KM, Kahn SE, Olefsky JM, Buchanan TA, and Scherer PE (2004) Complex distribution, not absolute am ount of ad iponectin, correlates with thiazolidinedione-mediated improvement in insulin sensitivity. J Biol Chem 279: 12152-12162. Park SH, Koh SS, Chun JH, Hwang HJ, and Kang HS (1999) Nrg1 is a transcriptional repressor for glucose repression of STA1 gene expression in Saccharomyces cerevisiae. Mol Cell Biol 19: 2044-2050.

PAGE 179

179 Patton JL and Lester RL (1991) The phosphoinositol sphingolipids of Saccharomyces cerevisiae are highly localized in the plasma membrane. J Bacteriol 173 : 3101-3108. Pennica D, Nedwin GE, Hayflick JS, Seeburg PH, Derynck R, Palladino MA, Kohr WJ, Aggarwal BB, and Goeddel DV (1984) Human tumo r necrosis factor: precursor structure, expression, and homology to lymphotoxin. Nature 312: 724-729. Petridou E, Mantzoros C, Dessypr is N, Koukoulomatis P, Addy C, Voulgaris Z, Chrousos G, and Trichopoulos D (2003) Plasma adiponectin concentr ations in relation to endometrial cancer: a case-control study in Greece. J Clin Endocrinol Metab 88 :993-997. Petridou E, Mantzoros C, Dess ypris N, Dikalioti SK, and Tric hopoulos D (2006) Adiponectin in relation to childhood myeloblastic leukemia. Br J Cancer 94: 156-160. Pinto WJ, Srinivasan B, Shepherd S, Schm idt A, Dickson RC, and Lester RL (1992) J Bacteriol 174: 2565-2574. Pyne S and Pyne NJ (2000) Sphingosine 1-phosphate signaling in mammalian cells. Biochem J 349: 385-402. Qi L, Rimm E, Liu S, Rifai N, and Hu FB (2005) Dietary glycemic index, glycemic load, cereal fiber, and plasma adiponectin concentration in diabetic men. Diab Care 28: 1022-1028. Qie L, Nagiec MM, Baltisberger JA, Lester RL and Dickson RC (1997) Identification of a Saccharomyces gene, LCB3 necessary for incorporation of exogenous long chain base into sphingolipids. J Biol Chem 272: 16110-16117. Rajala MW and Scherer PE (2003) Minireview: The adipocyteat the crossroads of energy homeostasis, inflammation, and artherosclerosis. Endocrinology 144: 3765-3773. Raas-Rothschild A, Pankova-Kholmyansky I, Kacher Y, and Futerman AH (2004) Glycosphingolipidoses: beyond the enzymatic defect. Glycoconj J 21: 295-304. Roberts WK and Selitrennikoff CP (1986) Isolatio n and partial characterization of two antifungal proteins from barley. Biochim Biophys Acta 880: 161-170. Robertson LS, Causton HC, Young RA and Fink GR (2000) The yeast A kinases differentially regulate iron uptake a nd respitory function. Proc Natl Acad Sci USA 97: 5984-5988. Rossler H, Rieck C, Delong T, Hoja U, and Sc hweizer E (2003) Functional differentiation and selective inactivation of multiple Saccharomyces cerevisiae genes involved in very-long-chain fatty acid synthesis. Mol Genet Genom 269: 290-298. Ross-MacD onald P, Sheehan A, Friddle C, Roeder GS, and Snyder M (1999) Transposon mutagenesis for the analysis of prot ein production, functi on, and localization. Methods Enzymol 303: 512-532.

PAGE 180

180 Rutherford JC and Bird AJ (2004) Metal-responsive transcription factors that re gulate iron, zinc, and copper homeostasis in eukaryotic cells. Eukaryot Cell 3: 1-13. Saba JD, Nara F, Bielawska A, Garrett S, and Hannun YA (1997) The BST1 gene of Saccharomyces cerevisiae is the sphingosine-1-phosphate lyase. J Biol Chem 272: 26087-26090. Sawai H, Okamoto Y, Luberto C, Mao C, Bielawska A, Domae N, and Hannun YA (2000) Identification of ISC1 (YER 019w) as inositol phosphosphin golipid phospholipase C in Saccharomyces cerevisiae. J Biol Chem 275: 39793-39798. Scherer PE, Williams S, Fogliano M, Baldini M and Lodish HF (1995) A novel serum protein similar to C1q, produced exclusively in adipocytes. J Biol Chem 270: 26746-26749. Schmidt MC and McCartney RR (2000) beta-subunits of Snf1 kinase are required for kinase function and substrate definition. EMBO J 19: 4936-4943. Schorling S, Vallee B, Barz WP, Riezman H, and Oesterhelt D (2001) Lag1p and Lac1p are essential for the acyl-CoA-dependent ceramide synthase reaction in Saccharomyces cerevisiae Mol Biol Cell 12: 3417-3427. Shenhar G and Kassir Y (2001) A positive regulato r of mitosis, Sok2, functions as a negative regulator of meiosis in Saccharomyces cerevisiae Mol Cell Biol 21: 1603-1612. Sims KJ, Spassieva SD, Voit EO, and Obeid LM (2004) Yeast sphingolipid metabolism: clues and connections. Biochem Cell Biol 82: 45-61. Sitruk-Ware R and Plu-Bureau G (1999) Progestins and Cancer. Gynecol Endocrinol 4: 3-9. Smith A, Ward MP, and Garrett S (1998) Msn2p /Msn4p-dependent gene expression to regulate growth, stress response and glycogen accumulation. EMBO J 17: 3556-3564. Spiegel S and Milstien S (2003) Sphingosine-1-phosphate: an enigmatic signaling lipid. Nat Rev Mol Cell Biol 4: 397-407. Stearman R, Yuan DS, Yamaguchi -Iwai Y, Klausner RD, and Dancis A (1996) A permeaseoxidase complex involved in highaffinity iron uptake in yeast. Science 271 : 1552-1557. Strum JC, Swenson KI, Turner JE, and Bell RM (1995) Ceramide triggers meiotic cell progression in Xenopus oocytes. A potential mediator of progesterone-induced maturation. J Biol Chem 270: 13541-13547. Sturgeon CM, Kemmer D, Anderson HJ, and Roberge M (2006) Yeast as a tool to uncover the cellular targets of drugs. Biotechnol J 1: 289-298.

PAGE 181

181 Tang YT, Hu T, Arterburn M, Boyle B, Bright JM, Emtage PC, and Funk WD (2005) PAQR proteins: a novel membrane receptor family defined by an ancient 7-transmembrane pass motif. J Mol Evol 61: 372-380. Thevelein JM (1994) Signal transduction in yeast. Yeast 10 : 1753-1790. Thompson-Jaeger S, Francois J, Gaughran JP, and Tatchell K (1991) De letion of SNf1 affects the nutrient response of yeast a nd resembles mutations which ac tivates the adenylate cyclase pathway. Genetics 129: 697-706. Toke DA and Martin CE (1996) Isolation and ch aracterization of a gene affecting fatty acid elongation in Saccharomyces cerevisiae. J Biol Chem 271: 18413-18422. Tomas E, Tsao TS, Saha AK, Murrey HE, Zhang CC, Itani SI, Lodish HF, and Ruderman NB (2002) Enhanced muscle fat oxidation and gl ucose transport by ACRP30 globular domain: acetyl-CoA carboxylase inhibition and AMPactivated protein kinase activation. Proc Natl Acad Sci USA 99: 16309-16313. Treger JM, Magee TR, and McEntee K (1998) Func tional analysis of the stress response element and its role in the multistress response of Saccharomyces cerevisiae Biochem Biophys Res Commun 243 : 13-19. Trujillo ME and Scherer PE ( 2005) Adiponectinjourney from an ad ipocyte secratory protein to biomarker of the metabolic syndrome. J Intern Med 257:167-175. Trujillo ME and Scherer PE ( 2006) Adipose tissue-derived factors: impact on health and disease. Endocr Rev 27: 762-778. Tsao TS, Murrey HE, Hug C, Lee DH, and Lodish HF (2002) Oligimerization state-dependent activation of NF-kappa B signaling pathway by ad ipocyte complement-related protein of 30 kDa (Acrp30). J Biol Chem 277: 29359-29362. Uetz P, Giot L, Cagney G, Mansfield TA, Judson RS, Knight JR, Lockshon D, Narayan V, Srinivasan M, Pochart P, Qureshi-Emili A, Li Y, Godwin B, Conover D, Kalbfleisch T, Vijayadamodar G, Yang M, Johnston M, Fields S and Rothberg JM (2000) A comprehensive analysis of protein-protein interactions in Saccharomyces cerevisiae. Nature 403: 623-627. Valentine JS and Gralla EB (1997) Deliver ing copper inside yeast and human cells. Science 278 : 817-818. Van Dyk D, Pretorius IS, and Bauer FF (2005) Mss11p is a central elem ent of the regulatory network that controls FLO11 express ion and invasive growth in Saccharomyces cerevisiae. Genetics 169 : 91-106. Van Ho A, Ward DM, and Kaplan J (2002) Transition metal transport in yeast. Annu Rev Microbiol 56 : 237-261.

PAGE 182

182 Vanhaesebroeck B and Alessi DR (2000) The PI3K-P DK1 connection: more than just a road to PKB. Biochem J 346: 561-576. Velculescu VE, Zhang L, Zhou W, Vogelstein J, Basrai MA, Bassett DE, Hieter P, Vogelstein B, and Kinzler KW (1997) Characterization of the yeast transcriptome. Cell 88: 243-251. Vincent O, Townley R, Kuchin S, and Carlson M (2001) Subcellular loca lization of the Snf1 kinase is regulated by speci fic beta subunits and a novel glucose signaling mechanism. Genes Dev 15: 1104-1114. Vyas VK, Kuchin S, and Carlson M (2001) Inte raction of repressors Nrg1 and Nrg2 with the Snf1 protein kinase in Saccharomyces cerevisiae Genetics 158 : 563-572. Vyas VK, Berkey CD, Miyao T, and Carlson M (2005) Repressors Nrg1 and Nrg2 regulate a set of stress-responsive genes in Saccharomyces cerevisiae Eukaryot Cell 4:1882-1891. Wang E, Norred WP, Bacon CW, Riley RT, and Me rill AH Jr (1991) Inhibition of sphingolipid biosynthesis by fumonisins. Implicat ion for diseases associated with Fusarium moniliforme J Biol Chem 266: 14486-14490. Wells GB, Dickson RC, and Lester RL (1998) Heat-induced eleva tion of ceramide in Saccharomyces cerevisiae via de novo synthesis. J Biol Chem 273: 7235-7243. Woods A, Munday MR, Scott J, Yang X, Carlson M, and Carling D (1994) Yeast SNF1 is functionally related to mammalian AMP-activated protein kinase and regulates acetyl-CoA carboxylates in vivo J Biol Chem 269:19509-19511. Wong GW, Wang J, Hug C, Tsao TS, and Lodish HF (2004) A family of Acrp30/adiponectin structural and functional paralogs. Proc Natl Acad Sci USA 101: 10302-10307. Wong J, Molyneaux L, Constantino MI, Tw igg SM, and Yue DK (2006) The metabolic syndrome in type 2 diabetes: When does it matter? Diabetes Obes Metab 8: 690-697. Xi W, Satoh H, Kase H, Suzuki K, and Hattori Y (2005) Stimulated HSP90 binding to eNOS and activation of the PI3-Akt pathway contribute to globular adiponectin-induced NO production: vasorelaxation in response to globular ad iponectin. Biochem Bi ophys Res Commun 332: 200205. Yamaguchi-Iwai Y, Stearman R, Dancis A, and Klausner RD (1996) Iron-regulated DNA binding by the AFT1 protein controls the iron regulon in yeast. EMBO J 15: 3377-3384. Ya mauchi T, Kamon J, Monokoshi Y, Ito Y, Waki H, Uchida S, Yamashita S, Noda M, Kita S, Ueki K, Eto K, Akanuma Y, Froguel P, Foufelle F, Ferre P, Carling D, Ki mura S, Nagai R, Kahn BB, and Kadowaki T (2002) Adiponectin stimulates glucose utilization and fatty-acid oxidation by activating AMP-activated protein kinase. Nat Med 8 : 1288-1295.

PAGE 183

183 Yamauchi T, Kamon J, Ito Y, Tsuchida A, Kita S, Sugiyama T, Miyagishi M, Ohteki T, Uchida S, Takekawa S, Waki H, Tsuno NH, Shibata Y, Te rauchi Y, Froguel P, Tobe K, Koyasu S, Taira K, Kitamura T, Shimizu T, Nagai R, and Ka dowaki T (2003) Cloning of adiponectin receptors that mediate antidiabetic metabolic effects. Nature 6941: 762-769. Zhao C, Beeler T, and Dunn T (1994) Suppressors of the Ca2+-sensitive yeast mutant (csg2) identify genes involved in sphingolipid bios ynthesis. Cloning and characterization of SCS1, a gene required for serine pa lmitoyltransferase activity. J Biol Chem 269: 21480-21488. Zhao H and Eide DJ (1996) The ZRT2 gene encodes the low affinity zinc transporter in Saccharomyces cerevisiae. Yeast 271 : 23203-23210. Zhao H and Eide DJ (1997) Zap1p, a metalloregu latory protein involved in zinc-responsive transcriptional regulation in Saccharomyces cerevisiae. Mol Cell Biol 17:5044-5052. Zhao H, Butler E, Rodgers J, Spizzo T, Duesterh oeft S, and Eide DJ (1998) Regulation of zinc homeostasis in yeast by binding of the ZAP1 transcriptional activat or to zinc-responsive promoter elements. J Biol Chem 44:28713-28720. Zhu B, Chen TH, and Li PH (1993) Expression of an ABA-responsive Osmotin-like gene during the induction freezing tolerance in Solanum commersonii Plant Mol Biol 21: 729-735. Zhu B, Chen TH, and Li PH (1995) Activation of two Osmotin-like pr otein genes by abiotic stimuli and fungal pathogen in transgenic potato plants. Plant Physiol 108 : 929-937. Zhu Y, Bond J, Thomas P (2003a) Identification, classification, and partia l characterization of genes in human and other vertabrates homologu es to fish membrane progestin receptor. Proc Natl Acad Sci USA 100: 2237-2242. Zhu Y, Rice CD, Pang Y, Pace M, Thomas P (2003b) Cloning expression, and characterization of membrane progestin receptor a nd evidence that is an intermediary in meiotic maturation of fish oocytes. Proc Natl Acad Sci USA 100: 2231-2236. Zimmet P, and Thomas CR (2003) Genotype,obesity and cardiovascular diseasehas technical and social advancement outstripped evolution? J Intern Med 254: 114-125.

PAGE 184

184 BIOGRAPHICAL SKETCH Brian Richard Kupchak was born in Scrant on, Pennsylvania on May 12, 1978. He grew up in northeastern Pennsylvania and graduated from Scranton High School in June of 1996. It was during th is time his love of science became a pparent from taking biology and human anatomy. He earned a B.S. in medical technology (2000) during his four years at the University of Scranton. It was here where he realized his true calling was in chemistry. During this time he also became certified as medical technologist and worked at Mo ses Taylor Hospital from 2000 to 2002. In 2002, he graduated from the University of Scranton with an M.A. in biochemistry. There, in the laboratory of Dr. Timothy Fole y, he studied the Effects of Reactive Nitrogen Species on Brain Protein Phosphatases. He fu rther enhanced his education by pursuing his Ph.D. in chemistry, beginning in August 2002, at the University of Florida in the laboratory of Dr. Thomas Lyons. Upon completion he will pursue a post-doctoral fellowship in sports nutrition at the University of Connectic ut in laboratory of Dr. Jeffery Volek.