OXIDANT-INDUCED CELL DEATH MEDIATED BY A Rho GTPase IN Saccharomyces cerevisiae

DISSERTATION

Presented in Partial Fulfillment of the Requirements for

the Degree Doctor of Philosophy in the Graduate

School of The Ohio State University

By

Komudi Singh

******

The Ohio State University 2008

Dissertation Committee: Approved by Dr. Hay-Oak Park, Advisor

Dr. Stephen Osmani ______Dr. Anita Hopper Advisor Graduate Program in Molecular Genetics Dr. Harold Fisk

ABSTRACT

Rho GTPases comprise a subgroup of the Ras superfamily of GTPases that controls a wide range of cellular processes by participating in processes such as actin cytoskeleton organization, phagocytosis, cell adhesion, cell-cycle progression and cell survival (Reviewed by Ridley, 2001). In budding yeast, Rho5 is a less characterized member of the Rho family of GTPase, and it shares considerable homology (45%) with

Rac GTPase (another member of the Rho family of GTPases). One of the better- established functions of Rac is activation of NADPH oxidase in neutrophils that is required for killing microbes (Abo et al., 1991). The Rho family of proteins have a unique “Rho insert region” and this domain has been implicated in Rac mediated

NADPH oxidase activation (Freeman et al., 1996). Historically, these oxidase were thought to function specifically in immune cells. However, in recent years, studies have shown that superoxides and ROS (reactive oxygen species) have important roles as signaling molecules, and they regulate a number of biological processes (reviewed by

Werner, 2004, Veal et al., 2007). Although there are a considerable number of reports on the events associated with the oxidative stress response, and the role of Rho GTPases in the production of ROS, the mechanism of sensing and oxidant-mediated signaling remains to be fully established.

ii In this study, the role of the Rho5 GTPase in hydrogen peroxide (H2O2)-induced cell death was explored. The presence of functional Rho5 in cells was found to confer sensitivity to various oxidants, including H2O2. Cells lacking Rho5 (i.e. Rho5-GTP bound form) exhibited much less cell death when exposed to H2O2, suggesting the presence of a

Rho5-mediated cell death program. This Rho5-mediated cell death process was triggered by H2O2 and exhibited some hallmarks of mammalian apoptosis such as ROS accumulation and DNA fragmentation. Additionally, Rho5 was also found to interact with Trr1 (Thioredoxin reductase), an important component of the thioredoxin antioxidant system. Interestingly, Rho1 GTPase another member of the Rho family of

GTPases was found to facilitate cell survival upon exposure to H2O2. These observations suggest that Rho GTPases control different cell fates depending on the oxidative stress level in promoting cell survival or cell death. This study uncovers a new role of a Rho

GTPase in oxidant-induced cell death in yeast. This work also provides evidence for

Rho5 interaction with Trr1, which is a novel link between a Rho GTPase and an antioxidant component.

This study raises important questions about the extent to which the apoptotic cell death machinery is conserved from yeast through higher multicellular systems, and also about the benefits that a unicellular organism like yeast gains by retaining this function.

This study has opened avenues to explore the above possibilities and hopefully identify more intracellular players functioning in the oxidative stress response and cell death processes. Identification of players that have homologous partners in higher systems will help determine the extent to which these cellular processes are conserved from yeast to humans.

iii DEDICATION

This work is dedicated to my family.

iv ACKNOWLEDGMENTS

I am deeply indebted to my advisor Dr. Hay-Oak Park for her continual guidance, and training throughout the years of my graduate studies. This achievement would not have been possible without her patience and motivation.

I would also like to thank my committee members, Dr. Stephen Osmani, Dr.

Anita Hopper, Dr. Harold Fisk and Dr. Berl Oakley for their time and guidance during my graduate studies at The Ohio State University.

I would like to thank Dr. Pil Jung Kang for his valuable guidance throughout the years of my graduate studies. Lastly, I would like to thank all the current and former members of the Park lab, for sharing their knowledge and resources during my stay in the lab.

v VITA

1997-2000……………………………..B.S. Nutrition, Food Service Management & Dietetics, University of Madras.

2000-2002…………………………….M.S. Biochemistry University of Madras.

2003-2005……………………………..Teaching Asst. The Ohio State University.

2005-2008……………………………..Reasearch Asst. The Ohio State University.

PUBLICATIONS

Singh, K., Kang, P.J., Park, HO. 2008. The Rho5 GTPase is necessary for oxidant- induced cell death in budding yeast. Proc. Natl. Acad. Sci. 105 (5): 1522-1527.

FIELDS OF STUDY

Major Field: Molecular Genetics

vi TABLE OF CONTENTS

ABSTRACT…………………………………………………………………………. ii DEDICATION………………………………………………………………………. iv ACKNOWLEDGEMENTS…………………………………………………………. v VITA………………………………………………………………………………… vi LIST OF TABLES…………………………………………………………………... ix LIST OF FIGURES…………………………………………………………………. x LIST OF ABBREVIATIONS……………………………………………………….. xiv

CHAPTERS:

1. INTRODUCTION…………………………………………………………. 1

G Proteins……………………………………………………………………. 1 Structures and properties………………………………………... 1 GTPase as molecular switches…………………………………… 3 Localization……………………………………………………… 4 Rho family of GTPases…………………………………………... 6

Oxidative stress and antioxidant systems………….………………………… 7

GTPases and oxidant-mediated cell signaling….…………………………… 9

Apoptosis and Necrosis……………………………………………………… 10

Yeast as a model system…………………………………………………….. 13

2. Rho5 IS NECESSARY FOR H2O2-INDUCED CELL DEATH………… 23

Introduction…………………………………………………………………. 23 Materials and Methods……………………………………………………… 24 Results………………………………………………………………………. 32 Summary……………………………………………………………………. 42

vii 3. Trr1 IS A POSSIBLE TARGET FOR Rho5……………………………... 66

Introduction………………………………………………………………….. 66 Materials and Methods………………………………………………………. 67 Results……………………………………………………………………….. 76 Summary…………………………………………………………………….. 83

4. DISCUSSION AND FUTURE DIRECTIONS…………………………… 100

How does Rho5 function in H2O2-induced cell death process?……………... 101 Why have cells retained the Rho5-mediated cell death program?…………… 111 Yeast and Apoptosis?.………………………………………………………... 112

Appendix A TEST THE ROLE OF Rho1-GTPase IN OXIDATIVE STRESS RESPONSE IN YEAST………….………………………………. 114

Appendix B TESTING THE BENEFITS OF Rho5- MEDIATED CELL DEATH IN AGING CULTURES………………………………... 127

Appendix C A CANDIDATE APPROACH TO IDENTIFY DOWNSTREAM TARGETS OF Rho5……………………………………………… 131

REFERENCES…………………………………………………………………….... 148

viii LIST OF TABLES

Table Page

Table 1. Strains used in this study…………………………………………… 140

Table 2. Plasmids used in this study………………………………………… 142

Table 3. Primers used in this study………………………………………….. 145

ix LIST OF FIGURES

Figure Page 1.1 Structure of small molecular weight GTPases………………………….……… 18

1.2 C-terminal structures and post-translational modifications of small GTPases………………………..…………………………………………….…. 18

1.3 CaaX modification………………………..…………………………….….…… 19

1.4 The Rho GTPase cycle………………………..………………………………... 20

1.5 Comparison of sequence of of Rho5 with other Rho and Ras family GTPases………………………..…………………………………..…………….. 21

1.6 Comparison of sequence of the carboxy-terminal region of Rho5 and Rsr1………………………..………………………..………………….………… 21

1.7 The thioredoxin antioxidant system…………………………………….……….. 22

2.1 Phenotype of rho5 mutants (in the presence of H2O2)…………………...……… 43

2.2 Phenotype of rho5Δ mutant expressing Rho5WT and mutant proteins from a plasmid………………………..………………………..…………………….….. 43

2.3 Phenotype of rho5 mutants (in the presence of paraquat and DEM)………….… 44

2.4 Survival of WT and rho5 mutants after H2O2 treatment……………………….… 45

G12V 2.5 Apoptotic phenotype of WT and rho5 after H2O2 treatment (ROS accumulation after DHR staining)………………………..………………….….. 46

G12V 2.6 Apoptotic phenotype of WT and rho5 after H2O2 treatment (Flow cytometry after DHR staining)………………………..………………………….….……… 47

G12V 2.7 Apoptotic phenotype of WT and rho5 after H2O2 treatment (DAPI staining)………..………………………..………………………………..……… 48

x G12V 2.8 Apoptotic phenotype of WT and rho5 after H2O2 treatment (TUNEL 50 staining)………………………..………………………..………………………..

2.9 Localization of WT Rho5-GFP fusion proteins…………………………………. 51

2.10 Localization of Rho5G12V-GFP fusion proteins…………………………………. 52

2.11 Sequence comparison of C-terminal region of Rho5 with other Rho and Ras GTPases………………………..………………………………………………… 53

2.12 Localization of CaaX mutant Rho5-GFP fusion proteins……………………….. 54

2.13 Localization of polybasic mutant Rho5-GFP fusion proteins…………………… 54

2.14 Growth phenotype of cells expressing Rho5C328S-GFP and Rho5K321-325S-GFP fusion protein………………………..…………………………………………... 55

2.15 Localization of Rho5-RitC-GFP fusion protein…………………………………. 56

2.16 Growth phenotype of cells expressing Rho5-RitC-GFP fusion protein…………. 57

2.17 Survival of cells expressing Rho5-RitC-GFP fusion protein after H2O2 treatment………………………..………………………………………………... 58

2.18 Staining of the vacuole prior to H2O2 treatment………………………………… 59

2.19 Staining of the vacuole after H2O2 treatment…………………………………… 60

2.20 Uptake of FM4-64 dye by endocytosis prior to H2O2 treatment………………… 61

2.21 Uptake of FM4-64 dye by endocytosis after H2O2 treatment…………………… 62

2.22 Staining of the vacuole prior to and after H2O2 treatment………………………. 63

2.23 Uptake of FM4-64 dye by endocytosis prior to and after H2O2 treatment………. 64

2.24 Methylene blue staining prior to and after H2O2 treatment……………………… 65

2.25 Testing the growth of rho5G12V mutant in the presence of non-fermentable carbon……………………………………………………………………………. 65

3.1 Interaction between Rho5 and Trr1 in vitro……………………………………... 85

3.2 Yeats two-hybrid assay………………………..………………………………… 86 xi 3.3 Interaction between Rho5 and Trr1 in vivo (Yeats two-hybrid assay)…………... 87

3.4 Bimolecular fluorescence complementation assay (BiFC)……………………… 88

3.5 Interaction between Rho5 and Trr1 in vivo (BiFC)…………………………….. 89

3.6 Immunoblot to detect YFPC fusions to the WT and mutant Rho5 proteins…….. 90

3.7 Elevation of Trr1-mCherry level in rho5Δ after H2O2 treatment………………... 92

3.8 Time course of Trr1-mCherry levels 0.5 h, 1 h, 2 h and 4 h H2O2 treatment……. 94

3.9 Localization and fluorescence intensity plot of Trr1-mCherry and Vph1-GFP prior to and after H2O2 treatment.…………………...…………………………… 96

3.10 Elevation of Trr1-mCherry level in rho5Δ after H2O2 treatment (Flow cytometry)……………………………………………………………………… 97

3.11 Trr1-HA levels in WT and rho5 mutants (western blotting)……….…………… 97

3.12 Growth phenotype of WT and rho5 mutants after treatment with 3mM H2O2…. 98

3.13 Northern analysis to test TRR1 mRNA levels in WT and rho5 mutants after H2O2 treatment………………………………………………………….……… 98

3.14 TSA1 overexpression rescues the cell death phenotype in WT cells after H2O2 treatment…………………………………………………………………………. 99

4.1 Growth phenotype of some rho1ts mutants in the presence of DEM and paraquat…………………………………………………………………………. 122

ts 4.2 Growth phenotype of some rho1 mutants in the presence of H2O2…………… 123

4.3 Growth phenotype of bck1Δ and mpk1Δ mutants in the presence of H2O2…….. 124

4.4 Localization of GFP-Rho1……………………………………………………… 125

4.5 Growth phenotype of tus1Δ and some ABC transporter deletion mutants……… 126

5.1 Competition assay between wildtype and rho5Δ strain…………………………. 130

5.2 Competition assay between wildtype and gic2Δ strain…………………………. 130

xii 6.1 Growth phenotype of rho5Δ strain in BY4742 background..…………………… 135

6.2 Growth phenotype of wildtype strain (in BY4742 background) carrying the rho5G12V plasmid…………………………………………………………… 135

6.3 Growth phenotype of aif1Δ strain (in YEF473 strain background) carrying the rho5G12V plasmid………………………………………………………………. 136

6.4 Growth phenotype of yca1Δ strain (in YEF473 strain background) carrying the rho5G12V plasmid………………………………………………………………. 137

6.5 Growth phenotype of nuc1Δ strain (in YEF473 strain background) carrying the rho5G12V plasmid………………………………………………………………. 138

6.6 Growth phenotype of ste20Δ strain (in YEF473 strain background) carrying the rho5G12V plasmid………………………………………………………………. 139

xiii LIST OF ABBREVIATIONS

2D two-dimensional a.a. amino acid

AIF apoptosis inducing factor

ATP adenosine triphosphate

AU arbitrary unit

BiFC Bimolecular fluorescence complementation

BPS bathophenanthroline disulfonate

CAD caspase activated DNase

CMAC CellTracker blue CMAC

DAPI 4',6-diamidino-2-phenylindole

DEM diethyl maleate

DEPC diethylpyrocarbonate

DHR dihydrorhodamine 123

DNA deoxyribonucleic acid

°C degrees Celsius

EDTA ethylenediaminetetraacetic

FACS Fluorescence Activated Cell Sorting

xiv Fe iron

FR free radical

FTase farnesyltransferase

GAP GTPase activating protein

GDP Guanosine diphosphate

GEF Guanine nucleotide exchange factor

GFP green fluorescent protein

GGTase geranylgeranyltransferase

GRX glutaredoxin

GSH glutathione

GTP Guanosine triphosphate

HA haemagglutinin

HCl Hydrochloric acid

H2O2 hydrogen peroxide his histidine

ICMT isoprenlycysteine carboxy methyltransferase leu leucine

MAPK Mitogen activated protein kinase

NaCl sodium chloride

NOX NADPH oxidase

- O2 superoxide radical

OH. hydroxyl radical

ORF open reading frame xv PCR polymerase chain reaction

PHOX phagocytic NOX

Pkc1 Protein kinase C 1

PM Plasma membrane

PMSF phenylmethanesulphonylfluoride

PS phosphatidylserine

RCE1 RAS-converting enzyme 1

Rho Ras Homologue

RhoGDI Rho guanine nucleotide dissociation inhibitor

RNA ribonucleic acid mRNA messenger RNA

ROS reactive oxygen species

SD synthetic dextrose

SDS sodium dodecyl sulphate trp tryptophan

Trr thioredoxin reductase

TrxR human thrioredoxin reductase

Trx thioredoxin ts temperature sensitive

TUNEL Terminal deoxynucleotidyl transferase dUTP nick end labeling

UTR untranslated region

VAMP vesicle associated membrane protein

VM Vacuolar membrane xvi WT wild-type

Yca yeast caspase

YFP yellow fluorescent protein

YPD yeast extract, peptone, dextrose

xvii CHAPTER 1

INTRODUCTION

G proteins

Small molecular weight GTP binding proteins, also called small GTPases, constitute a superfamily of regulatory proteins that control a diverse array of cellular processes. They function as molecular switches cycling between GDP-bound inactive state and GTP-bound active state. They are able to perform this function because all these proteins have GTP/GDP-binding property and GTPase (GTP-hydrolysis) activity.

To date, more than 100 members of the superfamily have been identified. These

GTPases are each categorized into distinct families based on sequence and function, including the Ras, Rho, Rab, Arf and Ran proteins (Reviewed by Takai et al., 2001). In budding yeast Saccharomyces cerevisiae, there are four Ras family members, six Rho family members (Rho1-5 and Cdc42), eleven Rab family members, seven Sar1/Arf family members and two Ran family members (Takai et al., 2001).

Structure and properties

Sequence comparison of G proteins has revealed considerable homology both at the primary structure level and amino acid sequence level. The regions responsible for

1 binding to GTP and GDP or for GTPase activity are often conserved. Binding of GTP results in a conformational change that will enable specific regions to bind other proteins and change their activity. These regions are called as the effector binding region and the proteins that bind to GTPases are the downstream effectors. In this way the GTPases are able to transduce an upstream signal to downstream effectors. Interestingly, considerable amino acid homology is found in the region responsible for binding to downstream effectors, between the GTPases (Figure. 1.1).

Among the Ras family of proteins, each protein shares relatively high (50-55%) amino acid identity, whereas Rab and Rho proteins (Rho, Rac and Cdc42) share 30% amino acid identity with Ras proteins. In addition to the amino acid homology, crystal structure analysis of several G proteins such as H-Ras, N-Ras, RhoA, Rac1 Rabs3A, and

Ran suggests that their GTP/GDP-binding domains have a common topology (Geyer and

Wittinghofer, 1997). Ras, Rho and Rab proteins have consensus sequences at the C- terminus that undergoes post-translational modifications necessary for membrane anchoring. The C-terminal region has been classified into at least 4 groups: Cys-a-a-X (or

CaaX) (a, aliphatic amino acid; X, any amino acid); Cys-a-a-Leu/Phe; Cys-x-Cys; and

Cys-Cys (Casey and Seabra, 1996, Glomseth and Farnsworth, 1994) (Figure. 1.2). The

Cys-a-a-X structure is further classified into two subgroups. One subgroup has an additional Cys residue upstream of the CaaX, and the other subgroup has a polybasic region. The CaaX structure is first farnesylated at the cysteine residue followed by proteolytic cleavage of aaX and carboxymethylation of the cysteine. GTPases like H-Ras that have an additional cysteine residue are palmitoylated. Lipid modifications of other forms of C-terminal structures is briefly summarized in Figure. 1.2. 2 The post-translational modification of the C-terminal CaaX is initiated by attachment of either a farnesyl (15-carbon long) or geranylgeranyl (20-carbon long) isoprenoid molecule (Winter-Vann and Casey, 2005). This reaction is catalyzed by farnesyltransferase (FTase) or geranylgeranyltransferase (GGTase-I), respectively. This is followed by the removal of the aaX tripeptide by a prenyl-protein-specific protease called Rce1 (Ras-converting enzyme 1). Finally, a methyl group is transferred to the cycteine by the enzyme isoprenlycysteine carboxy methyltransferase (ICMT) (Figure.

1.3).

GTPase as molecular switches

Structural analysis has revealed two interconvertible forms of GTPases: a GDP- bound inactive state and a GTP-bound active state (Figure. 1.4) (Hall et al., 1990). An upstream activation signal stimulates the release of GDP, which is followed by the binding of GTP, and eventually leads to a conformational change of the protein. This conformational change facilitates binding to downstream effector/s and signal transduction components. The GDP/GTP exchange reaction is catalyzed by regulators called guanine nucleotide exchange factors (GEF). Inactivation is achieved by the hydrolysis of the bound GTP to GDP through intrinsic GTP hydrolysis activity of the proteins (hence the term GTPase). The GTP hydrolysis activity of these proteins is enhanced by regulators called GTPase activating proteins (GAP) (Figure. 1.4).

Many studies have identified more than one GEF or GAP for a given GTPase, suggesting the presence of complex regulatory mechanisms. It is widely accepted that the

3 presence of more than one GEF or GAP would enable regulation of a GTPase in more than one cellular process. Consistent with this idea, one study in S. pombe has identified two regulators, Ste6 and Efc25, which regulate Ras1 GTPase in the processes of pheromone signaling and morphogenesis, respectively (Onken et al., 2006).

Localization

Studies in many labs have shown that lipid modifications are necessary for the membrane anchoring of GTPases. Many Ras and Rho proteins are synthesized as cytosolic proteins, but gain the capacity to associate with membranes after undergoing post-translational modifications described above. These proteins (and other GTPases) have been shown to be dynamically distributed in various compartments of the cells. For example, Ran localizes to both cytosol and nucleus (Takai et al., 2001). K-Ras localizes to the plasma membrane (PM) only, while H-Ras localizes to the PM and Golgi. TC10, a

Rho GTPase localizes to the endosomal compartments (Michaelson et al., 2001).

Unlike the Ras proteins, Rho proteins in mammalian cells have been shown to associate with RhoGDI (guanine nucleotide dissociation inhibitor), which inhibits the release of GDP from Rho proteins (Fukomoto et al., 1990). RhoGDI has been shown to extract inactive GTPases from the membrane. This association enables them to cycle on and off membranes, which adds more complexity to their localization and function.

Sequence comparison of the C-terminal end of some Rho and Ras GTPases have revealed second membrane targeting sequences present upstream of the CaaX motif.

RhoB and TC10, unlike H-Ras, have polybasic regions upstream of the CxxC motifs.

4 While the CxxC motif undergoes palmitoylation (Hancock et al., 1989), which regulates association with Golgi, the presence of polybasic region mediates association with endosomes (Michaelson et al., 2001). In contrast to RhoB, Rac1 has a polybasic stretch upstream of the CaaX motif, and is found to localize predominantly on the PM, suggesting that this signal engages in a trafficking pathway that does not allow accumulation in the internal membranes (Michaelson et al., 2001).

Similar localization studies in budding yeast with Rsr1/Bud1, a Ras GTPase, have revealed distinct roles for the CaaX motif and the polybasic motif in localizing the protein to the PM and internal organelles (Park et al., 2002). Rsr1/Bud1 with a functional

CaaX and polybasic motif localizes to the PM and to the vacuolar membrane. This study showed that mutation in the CaaX motif completely abolishes the membrane association of Bud1. In contrast, mutation of the polybasic residues abolishes the PM localization of

Bud1, but does not affect its localization on the vacuolar membrane. Similarly, the polybasic motif of Cdc42, a Rho GTPase, was found to be necessary for plasma membrane localization of Cdc42 (Richman et al., 2002).

Despite the presence of a common CaaX motif in the Ras and Rho family of proteins, the above studies have revealed that the presence of secondary regulatory sequences and their subsequent modifications influence the localization and function of

GTPases. For many Rho family GTPases, the nature/function of these secondary sequences is not fully understood. However, the presence of such sequences supports the model of very complex intracellular regulation of these proteins.

5 Rho family of GTPases

Rho GTPases are a subfamily of the Ras superfamily of GTPases, which control a wide range of signal transduction pathways (Etienne-Manneville and Hall et al, 2002).

Although the mechanism of the Rho switch is straight forward, Rho GTPases have a wide range of regulators (~60 GEFs and ~70 GAPs) that add complexity to its function and regulation (Etienne-Manneville and Hall, 2002).

Initial studies have revealed the role of some Rho GTPases in the assembly of distinct actin structures in a wide variety of mammalian cells. For example actin-rich protrusive structures formed at the leading edge of a motile cell are called filopodia and lamellipodia. Many cells also form contractile structures composed of actin and myosin bundles called stress fibers. Many studies showed that Rho, Rac, and Cdc42 regulate separate signal transduction pathways, resulting in the assembly of stress fibres, lamellipodia, and filopodia, respectively (Ridley and Hall, 1992, Nobes and Hall, 1995).

Many studies have also revealed their participation in other cellular processes such as cell polarity establishment, gene transcription, vesicular transport, and a number of enzymatic activities such as activation of NADPH oxidase in phagocytes and glucan synthase in yeast.

In immune cells, such as neutrophils and phagocytes, Rac1 GTPase was identified as an essential activator of the multi-component NADPH oxidase (Abo et al., 1991).

Activation of NADPH oxidase results in high levels of ROS production, which kills the pathogens engulfed by the immune cells. Surprisingly, there was no obvious homologue of Rac in budding yeast until the completion of yeast genome sequencing, which identified the Rho5 GTPase based on its sequence homology to other Rho GTPases. 6 Sequence analysis of Rho5 showed that it shares around 45% sequence homology with

Rac, and appears to be a unique Cdc42/Rac-like protein in budding yeast. The Ras-like effector region (a.a.32-40) of Rho5 is identical to that of Cdc42 and is quite similar to

Rac (Figure 1.5). In addition to this, Rho5 has an extra long amino acid sequence domain

(81 a.a.) near the C-terminus, which is similar to Rsr1 GTPase (a Ras GTPase involved in bud site selection). At present, it is not known whether this 81 a.a. region has any particular role (Figure 1.6).

Oxidative stress and antioxidant systems

All aerobic cells are under the constant threat of exposure to ROS that are either produced as a byproduct of cellular metabolism or in response to environmental stresses.

In broad terms, ROS refer to partially reduced forms of molecular oxygen that are highly reactive in nature. The highly reactive nature of these compounds enables them to damage a wide-range of cellular constituents such as DNA, lipids, and proteins.

- Additionally, superoxide radical (O2 ) serves as a reductant of iron (Fe). Reduced Fe

. reacts with H2O2, which results in the generation of hydroxyl radical (OH ) in a reaction called the Fenton reaction. Hydroxyl radicals are powerful oxidants that can cause massive cellular damage (Keyar and Imlay, 1989). As a result, cells have evolved defense mechanism to protect themselves from the damage caused by ROS (reviewed by

Jamieson, 1998).

Cells possess both enzymatic and non-enzymatic defense systems to protect themselves from oxidative damage and maintain cellular redox state. Non-enzymatic defense systems consist of small molecules that have the ability to scavenge free radicals

7 (FR). One of the best-known non-enzymatic defense systems is glutathione (GSH), which is a tripeptide that has a redox-active sulphydryl group. GSH scavenges oxidants by reducing them and forming oxidized glutathione (GSSG) (reviewed by Jamieson, 1998).

In addition to glutathione, thioredoxins (encoded by TRX1, TRX2 and TRX3 in yeast) are small sulphydryl rich proteins, which act as a reductant for thioredoxin peroxidase (encoded by TSA1 in yeast), an enzyme known to breakdown H2O2 to water

(reviewed by Jamieson, 1998; Veal et al., 2007). The thioredoxin proteins contain a dithiol active site responsible for their reducing action. In addition to thioredoxin peroxidase, they reduce other substrates such as ribonucleotide reductase and oxidoreductase (Gan, 1991). Carotenoids, vitamin E, vitamin C, and ascorbic acid are small molecules that also belong to the non-enzymatic defense system (Veal et al., 2007).

Cellular antioxidant defenses also include several enzymes that are capable of removing free radicals and their byproducts, as well as repairing some of the damage caused by oxidative stress. Some examples of these enzymes include catalase, superoxide dismutase, glutathione reductase, glutathione peroxidase, thioredoxin reductase, and thioredoxin peroxidase (reviewed by Jamieson, 1998).

The main function of glutathione reductase (encoded by GLR1 in yeast) is the maintenance of a high reduced to oxidized ratio of glutathione in cells. The thioredoxin reductase (encoded by TRR1 and TRR2 in yeast) functions similarly to GLR1, but maintains a high concentration of reduced thioredoxin, which in turn participates in various cellular processes (Gan, 1991, Pedrajas et al., 1999). The reduced thioredoxin acts as a hydrogen donor to thioredoxin peroxidase (Tsa1), which functions in the

removal of H2O2 from the cells (Figure 1.7; Chae et al., 1994). Studies have revealed the

8 presence of two thioredoxin antioxidant systems in yeast, one functions in the cytosol

(comprising the Trr1 and its substrates Trx1 and Trx2) (Trotter and Grant, 2002), and the other functions in the mitochondria (comprising the Trr2 and its substrates Trx3)

(Pedrajas et al., 1999). As discussed in chapter 3 and chapter 4, I found Rho5 interacted with Trr1 in a GTP-dependent manner in budding yeast providing a novel link between a

GTPase and an antioxidant component.

GTPases and oxidant-mediated cell signaling

Several studies in mammalian cells have implicated Rac GTPase in the production of ROS in immune cells. Upon stimulation, Rac-GTP translocates to the PM along with four other cytosolic factors, resulting in the assembly of a fully active oxidase complex (reviewed by Werner, 2004). The fully functional NADPH oxidase produces high concentrations of superoxide in a process called respiratory burst, which is necessary for killing the invading pathogens. Historically, the function and the tight regulation of these oxidases were thought to be unique to immune cells. However, studies in the recent years have suggested that ROS has a role in pathological as well as physiological processes. This is supported by studies that have discovered non- phagocytic NADPH oxidase (NOX) in non-phagocytic cells. These studies have shown that NOX produces lower levels of superoxide (when compared to phagocytic NOX

[PHOX]) in response to different physiological stimuli. These studies have also revealed that NOX is present in a wide range of cells such as liver cells, vascular cells and epithelial cells of the digestive tract and kidney (reviewed by De Minicis and Brenner.

2007; Brandes and Kreuzer, 2005).

9 It is known that hydrogen peroxide (H2O2, one of the better known ROS) is produced in response to various stimuli including cytokines and growth factors and has been shown to participate in the regulation of a number of biological processes (reviewed by Veal et al., 2007). However, several questions still remain about the role of Rho

GTPase in oxidant-mediated cell signaling as well as the signal transduction pathways participating downstream of these oxidants. My work has uncovered a novel role of the

Rho5 GTPase in the H2O2-induced cell death process, using budding yeast

Saccharomyces cerevisiae as a model system (see chapter 2 and chapter 4).

Apoptosis and Necrosis

Cell death forms an essential part of normal animal development and function. It participates in different biological systems such as immune system, normal cell turnover and embryonic development. During development, cell death facilitates formation of body parts by removing cells that are no longer needed. For example, removal of tail from a tadpole, separating digits etc. These cells activate an intracellular death pathway and kill themselves in a controlled way, a process called programmed cell death (PCD) or apoptosis (reviewed by Raff, 1998 and Cohen, 1997). By contrast, cells can die accidentally in response to acute injury/infection. In this case, the process is uncontrolled and results in cell swelling and eventually bursting resulting in spilling its content over neighboring cells resulting in inflammation, a process called necrosis.

The idea that animal cells have a built-in cell death program was further supported by identification of genes that were dedicated to apoptosis and its regulation

(reviewed by Ellis et al., 1991). Many of these genes encoded a family of proteases that

10 have cysteine in their active site and cleave their target proteins at specific aspartate residue and become activated in apoptosis, called caspases. Caspases are produced as large precursor molecules (called procaspases), which are activated by undergoing cleavage at aspartate residues. During apoptosis, caspases are activated in an amplifying cascade, cleaving one another in a sequence of proteolytic events. Once activated caspases cleave specific proteins in the cell to help kill it quickly and neatly (reviewed by

Raff, 1998). For example they cleave a protein that binds the DNA-degrading enzyme in an inactive state, releasing the DNase to cleave the DNA in the nucleus (Enari et al.,

1998).

Apoptosis is characterized by a series of distinct morphological and biochemical alterations. Apoptotic cells exhibit nuclear alterations, compaction of cytosolic organelles, decrease in cell volume and alterations in plasma membrane such as plasma membrane blebbing and externalization of phosphatidylserine (PS) (reviewed by Cohen,

1997). Changes in the plasma membrane helps mark the cell surface so that the apoptotic cells are recognized and eaten by macrophages (reviewed by Raff, 1998). The nuclear alterations are often associated with internucleosomal DNA cleavage that results on formation of a DNA ladder on agarose gel electrophoresis and is considered as a biochemical hallmark of apoptosis. Some studies have shown that chromosomal DNA damage plays a primary role in cell death in various cell lines. However, there are conflicting reports suggesting DNA cleavage as an early or late event in apoptosis

(Higuchi and Matsukawa, 1997; Cohen, 1997). ROS such as hydrogen peroxide (H2O2),

- hydroxyl radicals (•OH) and superoxide anions (O2 ) has also been shown to damage chromosomal DNA by oxidative nucleic acid base modification and single and double

11 strand DNA breaks (reviewed by Higuchi, 2003). In addition to ROS, caspase activated

DNase (CAD) and translocation of endonuclease G from mitochondria to the nucleus also contributes in DNA fragmentation (Enari et al., 1998; Li and Wang, 2001).

Despite distinct morphological differences between the apoptotic and necrotic cells, recent studies suggest the presence of complex initiating events leading to either cell death fates. Studies have shown that many insults that induce apoptosis at low dose, induce necrosis at high dose (reviewed by Zong and Thompson, 2007). Necrotic cells exhibit swelling accompanied by loss of plasma membrane integrity and DNA degradation mediated by DNaseI (reviewed by Higuchi, 2003). Thus the TUNEL assay, which has been used to label the 3' ends of nicked or fragmented DNA in apoptosis using labeled dUTP and terminal deoxynucleotide transferase enzyme, can also stain the

DNaseI degraded DNA.

Several labs have described yeast apoptosis during aging, mating, or during exposure to toxins (Fabrizio et al., 2004; Herker et al., 2004; Madeo et al., 1997).

Presence of apoptosis in a unicellular system like yeast has been a controversial issue due to the doubts cast on the benefit/s this process provides to a unicellular organism.

However, several labs have shown that the presence of early death program to sieve out old damaged cells from the population of yeast cells can facilitate the better survival of younger/healthier yeast cells. These studies also showed that yeast cells undergoing such altruistic cell death for the better survival of overall yeast population, exhibit some hallmarks of ammmalian apoptosis such as DNA fragmentation, externalization of PS, and ROS accumulation (Fabrizio et al., 2004; Madeo et al., 1997). However, usage of these criteria to support the existence of apoptosis can be a sensitive issue. This is

12 because to what extent the events responsible for apoptotic morphology contribute to yeast cell death is not clear. Nevertheless, these studies do support the presence of cell suicide programs in unicellular systems. To what extent these processes are conserved from yeast through humans and whether these processes are caspase dependent/independent are some interesting questions that remain to be answered.

Yeast as a model system

Budding yeast Saccharomyces cerevisiae serves as an ideal model system to study various biological processes because of its ease of genetic manipulation, completely sequenced and annotated genome, and short generation time (~90 minutes in rich media).

Unlike many other microorganisms, yeast can survive in haploid as well as diploid state.

This has enabled isolation of recessive mutations from haploid cells conveniently, and has also allowed complementation tests in diploid cells. Development of efficient DNA transformation procedure has facilitated introduction of plasmids into yeast cells conveniently.

Plasmids introduced into yeast cells can be maintained as replicating molecules or by integrating into the yeast genome. Some of these are autonomously replicating plasmids that can be maintained in high-copy number or low-copy number inside the cells. While the integrative plasmids do not replicate autonomously, they can be integrated into the genome. Integration of these DNA fragments inside yeast cells occurs via recombination. If the transforming DNA fragments carry partially homologous sequences, they can be targeted to specific location in the yeast genome by the process of homologous recombination. The budding yeast system has been found to be particularly

13 suitable for integrating a DNA fragment by the process of homologous recombination, if the fragment carries homologous sequences at its flanking ends. This has led to the development of techniques for direct replacement of DNA fragment of interest (i.e. genetically engineered DNA) into normal chromosomal location. Thus a normal wildtype gene can be replaced with an altered allele, or can be disrupted by a selectable marker.

All yeast plasmids carry marker that allow the selection of tranformants carrying desired plasmids. Most commonly used yeast plasmids carry genes that will complement specific auxotrophic mutations commonly present in lab yeast strains. For example a plasmid carrying a functional URA3 gene (involved in uracil biosynthesis) will complement the ura3-52 mutation present in the lab yeast strain. Similarly there are other plasmids that carry other auxotrophic markers like a functional TRP1, LEU2 andHIS3 gene (involved in trypatophan, leucine and histidine biosynthesis, respectively) that can complement the respective auxotropic mutation present in the yeast strain.

Since yeast have a short generation time, mass production of cells is easy, and it is also relatively easy to purify large quantities of proteins. Efficient protocols to isolate

DNA, mRNA, and tRNA are available. It is also possible to isolate subcellular organelles such as intact nuclei and mitochondria from yeast cells. In addition to the above- mentioned reasons, one main factor that has made yeast a successful model system in biological research is the remarkable extent to which several biological processes are conserved between yeast and other higher systems.

In the recent years, yeast has emerged as an important model system in the development of current genomic and proteomic studies. After the completion of the yeast genome sequencing (Goffeau et al., 1996), several genome/proteome-wide studies have

14 enabled analysis of gene function, mechanism of gene regulation, and also identify network of genes participating in various biological processes (reviewed by Kumar and

Snyder, 2001). Some of these studies are mentioned below.

Transposons have long been used as mutagenic agents in traditional genetics studies. Several labs have used these transposable elements to generate gene disrucption or conditional mutations (that can be used to study the essential gene function).

Recombination using transposon-encoded HA epitope has been used to epitope tag several genes that can be immunolocalized or immunoprecipitated (Devine and Boeke,

1994; reviewed by Kumar and Snyder, 2001). One limitation for transposon-based mutagenesis is that the process is random and hence will be difficult to cover the entire genome. To overcome these limitations, a PCR-based approach was undertaken to systematically delete each annotated gene in the yeast genome (Winzeler et al., 1999).

The deletion collection generated from this work has been extensively used to characterize drug responses, drug targets (Chan et al., 2000), and has also enabled phenotypic analysis on a genome-wide scale. Another study generated a collection of yeast strains expressing full-length, chromosomally tagged green fluorescent protein fusion proteins enabling global analysis of protein localization (Huh et al., 2003).

Recently DNA microarray has been used to identify genes whose expression is either induced/repressed, or differentially expressed during a specific cellular response.

Microarrays has been extensively used in yeast to perform similar studies, for example to identify genes differentially expressed in cellular processes like during sporulation, shift from anaerobic fermentation to aerobic respiration (reviewed by Kumar and Snyder,

2001). Severals other studies have profiled gene expression during stress conditions,

15 chemical exposures, etc (Causton et al., 2001).

In the recent years, a genome-wide tagging of proteins have enabled development of protein microarray chips that has enabled genome-wide analysis of protein function.

These protein arrays have been used to analyze protein-protein interaction and enzyme- substrate specifity etc. Several high-throughput technologies utilize a combination of two-dimensional (2D) gel electrophoresis, peptide sequencing and mass spectrometry to resolve proteins present in large complexes (reviewed by Kumar and Snyder, 2001).

Recently many established experimental methods have been used on a genome- wide scale. For example, a yeast two-hybrid system has been extensively used to study protein-protein interactions. Several groups have systematically used yeast two-hybrid assays by tagging annotated yeast open reading frames (ORF) as bait or prey (in different combinations) to generate protein-protein interaction map (reviewed by Kumar and

Snyder, 2001). Additionally a systematic genetic analysis was performed to identify gene pairs that partivcipate in similar biological processes or function in a single protein complex (Tong et al., 2001). This approach identified several genetic interactions that were used to generate a global map for gene function.

Yeast as a model system to study apoptosis: Apoptosis is a form of programmed cell death that plays a central role in homeostasis and maintenance of metazoans. One of the key features of mammalian apoptosis is its effectiveness, i.e. the removal of cells happen rapidly, and without any damage to the nearby cells. Several studies have identified a complex network of regulators and effectors that participate in this process. To further add to the complexity, the regulators/effectors, and the regulatory pathways vary depending on the developmental state, tissue and cell type of the organism. All these

16 factors have made the molecular analysis of apoptosis process tricky, often resulting in contradictory models to explain the same process. A simple model system like yeast could be used to study these processes at the molecular level. However, apoptotic programmed cell death was assumed to be confined to the multicellular system until recently, when several labs showed the presence of PCD in many unicellular systems

(reviewed by Rockenfeller and Madeo, 2008).

One such study showed that, like metazoan cells, yeast also undergoes cell death displaying several apoptotic markers such as chromosomal condensation, DNA fragmentation, and exposure of PS to the exterior of the plasma membrane (Madeo et al.,

1997). This type of cell death in yeast was shown to be both endogenically triggered as well as triggered by an external agent like H2O2 and expression of mammalian pro- apoptotic gene (reviewed by Rockenfeller and Madeo, 2008). For several years, yeast has been used as a model to study replicative and chronological aging process. Studies have shown that both aging forms end in programmed necrotic or apoptotic cell death in yeast.

Rescue experiments using deletion of pro-apoptotic machinery or diminishing ROS suggested a connection between aging, oxidative stress and cell death. Additionally, several yeast orthologes of mammalian apoptotic regulators have been identified, such as yeast caspase (Yca1; Madeo et al., 2002), endoG homologue Nuc1 (Buttener et al.,

2007), Aif1 (Wissing et al., 2004), and Ste20 (Ahn et al., 2005). All these factors suggest the presence of a basic apoptotic machinery in yeast. Hence, using yeast as a model to study apoptosis will provide insights into the molecular mechnism underlying this process, and also provide clues about the extent to which this process is conserved from unicellular fungi through complex metazoans.

17 18 19 20 21 22 CHAPTER 2

Rho5 IS NECESSARY FOR H2O2-INDUCED CELL DEATH IN YEAST.

INTRODUCTION:

The redox balance is disturbed when yeast cells are exposed to diverse environmental stresses such as oxidants, heat shock, and metal ions. It has been reported

that H2O2 treatment triggers apoptotic cell death in yeast as it also does in various mammalian cells (Madeo et al. 1999, Cai. 2005), although to what extent the process of apoptosis is conserved from yeast to mammals is not clear {for reviews, see (Jin et al.

2002, Hardwick and Cheng, 2004, Buttner et al. 2006}. Small GTPases are involved in apoptotic and necrotic cell death (Kawasaki et al. 1999, Lacal. 1997, Vanden Berghe et al. 2007) as well as production of ROS in higher eukaryotes (Werner. 2004). In particular, Rac GTPase is involved in the activation of NADPH oxidase, which accepts electrons from NADPH to produce the superoxide radical in neutrophils and non- phagocytic cells (Werner, 2004, Hordijk, 2006). Rac GTPase is a member of the Rho family of GTPases.

In yeast, the Rho family of GTPases consists of 6 members: Rho1-5 and Cdc42. I am interested in Rho5 because of its unique structural features mentioned below. The N terminal region of Rho5 including the putative effector region is similar to that of Cdc42

23 and human Rac1 (45%homology) (Garcia-Ranea and Valencia, 1998), whereas its C terminal region is similar to that of Rsr1 (a Ras family GTPase required for bud site selection). These and other structural features suggest an interesting function and regulation of Rho5. However, the cellular role of Rho5 has been elusive as rho5 deletion mutants are viable and grow fine at any temperature. Given its structural similarity to

Rac, I asked whether Rho5 affects redox homeostasis in yeast. This chapter provides

evidence supporting that Rho5 is necessary for H2O2-induced cell death, which accompanies ROS accumulation and vesicular trafficking defects.

MATERIALS AND METHODS

Standard methods of yeast genetics, recombinant DNA manipulation, and growth conditions were used (Ausubel et al., 1999; Guthrie and Fink, 1991) unless indicated otherwise .

Yeast strains and their phenotype in the presence of oxidants

Yeast strains used in this study are listed in the Strain Table (Table 1).

Laboratory (wild type) WT strains exhibited varying degrees of sensitivity to H2O2 depending on the background. HPY210 (WT used in this study) was less sensitive to 1- 2

mM H2O2 than BY4742 (from Open Biosystems) - the WT strain in the background used by the Saccharomyces Genome Deletion Consortium. This difference may explain why trr1 deletion causes lethality in the BY4742 background, unlike in the HPY210 strain background. Despite the differences, the rho5Δ mutant in the BY4742 background was

more resistant to 1 mM H2O2 than BY4742 (see APPENDIX C). All strains used in this 24 study were isogenic to HPY210 except as indicated in Strain Table 1, and rho5 mutants

exhibited a distinct, reproducible phenotype upon exposure to 4 - 4.5 mM H2O2.

Growth phenotype and cell viability assay

WT and rho5Δ strains were diluted to OD600 = 0.4 from a mid-to-late logarithmic phase culture in YPD and then serially diluted (10-fold). These cells were spotted on

YPD plates containing 400µg/ml paraquat (Sigma-Aldrich Chemical Co), 1 mM diethyl maleate (DEM) (Sigma-Aldrich Chemical Co), or no oxidant. The rho5Δ strain carrying each RHO5 plasmid was tested similarly but in 2.5-fold serial dilutions using SC medium

(or plates) lacking uracil and 200µg/ml paraquat. The plates were incubated at 30°C for

2-7 days.

To test the growth phenotype in the presence of H2O2, cells carrying each RHO5 plasmid were grown to mid logarithmic phase (in SD-URA media) were then diluted to

OD600 = 1, and then serially diluted (5-fold). 200 µl of each culture was transferred to a

96-well plate in the presence or absence of 4.4mM H2O2 and incubated for 4hrs at 30°C.

The cells were plated on SD-URA plates using sterile metal pins. The growth phenotype

of WT and various rho5 mutants were also tested in the presence of H2O2 as described above, except that the cells were grown in YPD media (with or without H2O2) and were plated on YPD plates.

To test cell viability, cells were diluted to OD600 = 0.4 from the exponentially growing cultures. A portion of the diluted cells was grown in the presence of 4.4 mM

H2O2 for 4hrs or mock-treated. The number of viable cells was estimated by counting colonies formed from equal OD600 units of each culture plated prior to and after 4 hr 25 incubation in the presence or absence of H2O2. Cell viability is given by the ratio

(number of viable cells after 4 hr incubation / number of viable cells prior to H2O2 treatment) X 100. I also estimated total cell number by directly counting cells with a hemacytometer before plating, with similar results.

Determination of ROS Accumulation

ROS accumulation was monitored by treating cells with an oxidation-sensitive fluorescent dye dihydrorhodamine 123 (DHR) (Sigma Chemical Co.). Upon oxidation, this dye becomes fluorescent and cells exhibit red fluorescence. Cells from mid-log phase

cultures were diluted to OD600 = 0.3, and then incubated with H2O2 (4.4 mM final) for 4 hrs. This same condition of H2O2 treatment has been used for all the localization studies.

For the last 2 hrs of the incubation with H2O2, DHR was added to a final concentration of

5 µg/ml, then cells were observed under the fluorescence microscope with TRITC filter or subjected to flow cytometry using a FACS Calibur (Becton Dickinson, Mountain

View, CA) following the procedure of Madeo et al. (Madeo et al. 1999).

Staining and Microscopy

Staining with FM4-64 (Invitrogen) was performed for 40 min, as described by

Vida and Emr (Vida and Emr, 1995). To visualize nuclei, cells were stained with DAPI

(4',6-diamidino-2-phenylindole; 10mg/ml final) for 5-8 min, after H2O2 treatment.

Localization of GFP fusion proteins was determined as previously described (Kang et al.,

2001), except where indicated cells were treated with H2O2 prior to observing under the microscope. Image acquisition was carried out using a Nikon E800 microscope (Nikon, 26 Tokyo, Japan) fitted with a 100X oil-immersion objective (N.A.= 1.30) as previously described (Kang et al., 2001), except that Slidebook 4 digital microscopy software

(Intelligent Imaging Innovations Inc.) was used to capture a series of optical sections at

0.3-µ intervals. A representative single section from a Z-series of images was used to study the localization pattern of the proteins. Some Z-series images were processed with no neighbors deconvolution and a representative single section was shown where indicated.

Cloning, deletion, and tagging of RHO5

Plasmids and Primers used in this study are listed in the Plasmid Table (Table 2) and

Primer Table (Table 3), respectively.

To construct the RHO5 deletion, a DNA fragment (3.3 kb) carrying rho5Δ::kanMX4 was amplified by PCR using genomic DNA from HPY657 (the rho5Δ::kanMX4 strain obtained from Open Biosystems; confirmed by genomic PCR) as a template and primers oRHO51 and oRHO52. The PCR product was used to delete the

RHO5 gene in YEF473 (Bi and Pringle, 1996) by one-step gene disruption (Rothstein,

1991), yielding HPY713, which was verified by genomic PCR. Among the meiotic progeny of HPY713, rho5Δ::kanMX4 co-segregated with resistance to oxidants such as

DEM and paraquat (data not shown), yielding HPY720.

To sub-clone the RHO5 gene, the XmaI-SacI fragment carrying RHO5 was isolated from the genomic library plasmid pSG14-URA3 (a kind gift from Hithen

Madhani, University of California-San Francisco) and cloned into pRS426 (Christianson

27 et al., 1992), yielding pHP1280. The XmaI-SacI fragment of rho5G12V was isolated from pSG14V12-306 (a kind gift from Hithen Madhani, University of California-San Francisco), and cloned into pRS426, yielding pHP1281. The rho5K16N mutation was introduced into pHP1280 (by Stratagene site-directed mutagenesis) using primers oRHO55 and oRHO56, yielding pHP1282. Chromosomal integration of rho5G12V was performed by transforming

HPY720 with plasmid pSG14V12-306 after digestion with NheI, yielding HPY999.

To express Rho5-GFP, a domain near its C-terminus, which contains extra amino acid residues (a.a. 248-316) compared to other Rho GTPases, was replaced with GFP in two steps: First, to amplify DNA fragments encoding both sides of the domain but excluding a.a. 248-316, two PCR reactions were carried out using pHP1280 as a template and primers oRHO512 and oRHO516, yielding a 1.5-kb PCR product, and oRHO515 and oRHO52 , yielding a 461-bp PCR product. The 1.5-kb fragment, digested with XmaI and NotI (sites included in the primers), and 461-bp fragment, digested with NotI and

XhoI (sites included in the primers), were cloned into the XmaI-XhoI site of pHP1476 (a pRS426 derivative that lacks the NotI site), yielding pHP1544. Second, a 720-bp NotI fragment encoding GFPS65T, V163A, S175G, isolated from pHP767 (Park et al., 2002), was inserted into the NotI site of pHP1544 in the correct orientation, yielding pHP1499. To construct pRS306-RHO5-GFP, pHP1499 was digested with XmaI and XhoI, and the resulting 2.8-kb RHO5-GFP fragment was ligated into the XmaI-XhoI site of pRS306

(Sikorski and Hieter,1989), yielding the plasmid pHP1467. Chromosomal integration of

RHO5-GFP was performed by transforming HPY720 with pHP1467 after digestion with

SphI, yielding HPY1156. Rho5-GFP was confirmed to be functional based on testing the

28 growth phenotype of HPY1156 in the presence oxidants such as paraquat, diethyl

maleate, and H2O2 (data not shown).

To express Rho5G12V-GFP, DNA fragments carrying rho5G12V were amplified by

PCR using pHP1281 as a template and primers oRHO512 and oRHO516. The 1.5-kb

PCR product, digested with XmaI and NotI (sites included in the primers), was ligated into the XmaI-NotI site of pHP1544, yielding pHP1502. The GFP fragment was then introduced into the NotI site of pHP1502 as described above, yielding pHP1503. To construct pRS306- rho5G12V-GFP, the 2.8-kb XmaI-XhoI fragment from pHP1503 was cloned into pRS306 digested with XmaI and XhoI, yielding pHP1505. To express

Rho5G12V-GFP from the chromosomal RHO5 locus, a rho5Δ strain (HPY720) was transformed with pHP1505 after digestion with SphI, yielding HPY1157. Rho5K16N lacking a.a. 248-316 in pHP1476 was constructed by PCR using pHP1282 as a template and primers oRHO512 and oRHO516. The 1.5-kb PCR product, digested with XmaI and

NotI (sites included in the primers), was ligated into the XmaI-NotI site of pHP1544, yielding pHP1507.

To express Rho5-GFP carrying a mutation in the CaaX motif, primers were designed to introduce a Cysteine to Serine mutation that will also result in the generation of a XbaI site. Two PCR reactions were carried out using pHP1499 as template. First, a

1.548-kb PCR fragment encoding the Rho5 CaaX motif along with upstream sequences was amplified by primers oRHO512 and oRHO520, and a 466-bp PCR product encoding the Rho5 region downstream of the CaaX motif was amplified by primers oRHO519 and oRHO52. The resulting 1.54-kb PCR product digested with XmaI and XbaI (sites included in the primers) and the 466-bp PCR product digested with XbaI and XhoI, were 29 cloned into the XmaI-XhoI site of pHP1476 , yielding pHP1535. The GFP fragment was then introduced into the NotI site of pHP1502 as described above, yielding pHP1537. To construct pRS306-rho5C328S-GFP, pHP1537 was digested with XmaI and XhoI, and the resulting ~2.8-kb rho5C328S-GFP fragment was ligated into the XmaI-XhoI site of pRS306, yielding the plasmid pHP1539. To express rho5C328S-GFP from the chromosomal RHO5 locus, a rho5Δ strain (HPY720) was transformed with pHP1539 after digestion with SphI, yielding HPY1185.

To express Rho5-GFP carrying a mutation in the polybasic motif, primers were designed to introduce five Lysines to Serine mutations that will also result in the generation of a SacI site. Two PCR reactions were carried out using pHP1499 as a template. First, a 1.508-kb PCR fragment encoding the Rho5 region including three lysine residues and sequences upstream of the polybasic motif was amplified by primers oRHO512 and oRHO518, and a 489-bp PCR product encoding the Rho5 region downstream of the polybasic motif was amplified by primers oRHO517 and oRHO52.

The resulting 1.5-kb PCR product digested with XmaI and SacI (sites included in the primers) and 489-bp PCR product digested with SacI and XhoI, were cloned into the

XmaI-XhoI site of pHP1476, yielding pHP1536. The GFP fragment was then introduced into the NotI site of pHP1502 as described above, yielding pHP1538. To construct pRS306-rho5K321-325S-GFP, pHP1538 was digested with XmaI and XhoI, and the resulting

~2.8-kb rho5K321-325S-GFP fragment was ligated into the XmaI-XhoI site of pRS306, yielding the plasmid pHP1540. To express rho5K321-325S-GFP from the chromosomal

RHO5 locus, a rho5Δ strain (HPY720) was transformed with pHP1540 after digestion with SphI, yielding HPY1186. 30 To construct a Rho5-RitC fusion protein, 18 a.a. residues present at the C- terminal end of Rho5 (that includes the CaaX and the polybasic motif) was deleted and replaced with the C-terminal domain of RitC, which has 60 a.a residues. A 186-bp C- terminal domain of RitC was isolated from the plasmid pHP1530 by digesting the plasmid with SacI and XbaI. A 480-bp PCR fragment encoding the Rho5 region downstream of the stop codon was amplified by primers oRHO525 and oRHO52. The resulting PCR fragment was digested with SacI and XhoI (sites included in the primers).

The above PCR product and the RitC fragment were then cloned into the XbaI-XhoI site of pRS306, yielding pHP1669. Next, a domain of Rho5 near its C-terminus, which contains extra amino acid residues (a.a. 248-316) compared to other Rho GTPases, was amplified by PCR using primers oRHO523a and oRHO524a. The 198-bp PCR product was then digested with NotI and XbaI (sites included in the primer). The 647-bp fragment comprising the RitC domain and the C-terminal domain of Rho5 was isolated from pHP1669 by digesting the plasmid with XbaI and XhoI. The above-described two fragments were then cloned into the NotI-XhoI site of pHP1544, yielding pHP1670. The

GFP fragment was then introduced into the NotI site of pHP1670 as described above, yielding pHP1671. To construct pRS306-rho5-RitC-GFP construct, pHP1671 was digested with XmaI and XhoI, and the resulting ~3.1-kb rho5-RitC-GFP fragment was cloned into the XmaI-XhoI site of pRS306, yielding pHP1672. To express rho5-RitC-GFP from the chromosomal RHO5 locus, a rho5Δ strain (HPY720) was transformed with pHP1672 after digestion with SphI, yielding HPY1506.

31 TUNEL assays

TUNEL assays were performed essentially as described by Zhang et al. (Zhang et al., 2006) except for the following modifications. WT and rho5Δ cells were treated with

600 µg/ml Zymolyase 100T (ICN Biomedicals Inc.) for 40 min and 1hr 20 min, respectively. The rho5G12V cells were treated with 300 µg/ml Zymolyase 100T for 20 min.

RESULTS

A rho5 deletion mutant is resistant to oxidants.

To test whether Rho5 affects the redox balance in yeast, I examined the growth phenotype of rho5 mutants in the presence of oxidizing agents such as DEM, a thiol- specific oxidant that depletes glutathione in the cell, and paraquat, a superoxide- generating agent. Both of these reagents increase intracellular ROS levels (Carmel Harel et al., 2001, Alic et al., 2001, Wanke et al., 1999). Although both WT and a rho5Δ mutant grew at about the same rate in the absence of oxidants, rho5Δ grew better (about 10-fold) than WT on plates containing DEM or paraquat (Fig. 2.1).

To examine whether the sensitivity to oxidants depends on the guanine nucleotide-bound state of Rho5, I used a constitutively active rho5 mutant (rho5G12V), which carries a substitution of Gly at position 12 with Val. A constitutively inactive rho5 mutant gene (rho5K16N), which is expected to express the GDP-locked (or nucleotide- empty) Rho5 was also generated. The rho5Δ cells carrying each mutant or a WT RHO5 plasmid were tested for their growth on oxidant-containing plates. Although there was a relatively mild difference in sensitivity, the rho5Δ cells carrying the rho5G12V plasmid were more sensitive to the oxidants than the cells carrying the WT plasmid or vector 32 control, whereas the rho5Δ cells carrying the rho5K16N plasmid were more resistant to the oxidants than the cells carrying the WT plasmid (Fig. 2.2).

In addition to DEM and paraquat, H2O2 is a widely used oxidant to study oxidative stress response in cells. H2O2 has also been shown to trigger apoptotic-like cell death in budding yeast (Madeo et al., 1999, Ahn et al., 2005). Hence the growth

phenotype of rho5 mutants after H2O2 treatment was examined (see MATERIALS &

METHODS for details). Equal number of rho5Δ cells (serially diluted in 5-fold ratio)

carrying each of the RHO5 plasmids were treated with H2O2 in a 96-well plate and spotted on a SD-Ura plate. Consistent with the above observations, cells expressing

G12V Rho5 mutant protein were hypersensitive to H2O2, while the rho5Δ cell carrying the

empty plasmid were resistant to H2O2 (Fig. 2.3). Taken together, these results suggest that

Rho5-GTP is required for the oxidant-induced cell death of budding yeast.

Rho5 is required for H2O2-induced cell death.

Since studies in other labs had shown that external doses of H2O2 induced apoptosis in yeast, the cell viability of WT and rho5 mutant cells was determined after

H2O2 treatment. Consistent with previous reports, the WT strain exhibited about 30% viability after H2O2 treatment. Interestingly, the rho5Δ mutant exhibited much higher cell survival (~ 80%), whereas cells expressing Rho5G12V from the chromosomal locus

exhibited less cell survival (~ 10%) after H2O2 treatment (Fig. 2.4).

To test whether Rho5-dependent cell death is associated with apoptotic phenotypes such as ROS accumulation and DNA fragmentation, I first monitored the intracellular ROS level with dihydrorhodamine 123 (DHR), which accumulates in the cell 33 and becomes fluorescent rhodamine 123 upon oxidation (Herker et al., 2004). Before

exposure to H2O2, little fluorescence was detectable in WT and rho5 mutants (data not shown and Fig. 2.6). After H2O2 treatment, the fluorescent signal dramatically increased in the rho5G12V cells and to a lesser extent in WT, but little in the rho5Δ cells (Fig. 2.5).

Analysis of these cells by flow cytometry also indicated a high level of ROS in rho5G12V

after H2O2 treatment but not in rho5Δ, while little fluorescence was detected in all these

cells prior to H2O2 treatment (Fig. 2.6). I also examined the nuclei of H2O2 treated rho5 mutant cells with DAPI staining. The rho5Δ cells showed intact round nuclei as well as mitochondrial DNA, which appeared as smaller dots predominantly located near the

periphery of the cells, prior to and after H2O2 treatment (Fig. 2.7). In contrast, about 62% of the rho5G12V cells showed misshapen or fragmented nuclei and most of them lacked

G12V mitochondrial DNA staining after H2O2 treatment (Fig. 2.7). The rho5 mutant, however, exhibited normal nuclear and mitochondrial DNA staining and were able to

grow on the medium based on the non-fermentable carbon source prior to H2O2 treatment

(Fig. 2.25), suggesting that the mutant cells contain functional mitochondria.

These cells were also double-stained by TdT-mediated dUTP nick end labeling

(TUNEL) to detect DNA fragmentation, which is a hallmark of apoptosis. Although no

G12V TUNEL positive cells were found prior to H2O2 treatment, 87% of rho5 and 76% of

WT cells were TUNEL-positive after H2O2 treatment, whereas 23% of the rho5Δ cells were TUNEL-positive (Fig. 2.8). Taken together, these data suggest that Rho5 is

involved in the process of cell death induced by H2O2 that exhibits DNA fragmentation

(similar to mammalian cells undergoing apoptosis).

34 G12V Localization of Rho5-GFP and Rho5 -GFP before and after H2O2 treatment

To gain insight into how Rho5 is involved in H2O2-induced cell death the localization of Rho5 was examined using a functional GFP fusion expressed from its chromosomal locus (see Materials and Methods). The localization of Rho5G12V-GFP was also examined using a strain that is similar except for the rho5G12V mutation. Prior to

G12V exposure to H2O2, both Rho5-GFP and Rho5 -GFP localized to the plasma membrane and to the endomembranes, mainly the vacuolar membrane based on staining with the vital dye FM4-64 (Fig. 2.9 & 2.10, a-c), which is delivered to the vacuolar membrane via

an endocytic transport route (Vida and Emr, 1995). After H2O2 treatment, Rho5-GFP also appeared in the vacuolar lumen in addition to the membranes (25~60%) (Fig. 2.9, d-f), while Rho5G12V-GFP appeared in a punctate pattern (35~65%) (Fig. 2.10, d-f), but little on the plasma membrane. Most patches of Rho5G12V-GFP overlapped with the patches from FM4-64 staining (Fig. 2.10, d-f), suggesting that Rho5G12V-GFP localized to the endocytic (and/or secretory) vesicles.

Localization of Rho5 to the plasma membrane is likely to be necessary for H2O2- induced cell death

Since Rho5-GFP localized to the plasma membrane as well as the vacuolar membrane, I wanted to determine which membrane association is necessary for Rho5

function in H2O2–dependent cell death, and from which membrane Rho5 translocates into the vacuolar lumen. Rho5 has a CaaX (C is Cysteine; a is aliphatic; X is any amino acid) motif at the C-terminus and a polybasic motif consisting of five Lysine residues near the

C-terminus (Fig. 2.11). These sequences have been implicated with specific membrane

35 association in other GTPases. Thus I created Rho5C328S-GFP (by substituting Serine at position 328 for Cysteine) and Rho5K321-325S-GFP (by substituting Serine at positions 321,

322, 323, 324, and 325 for Lysines) (Fig. 2.11). As I expected, Rho5C328S-GFP localized diffusely to the cytosolic compartment (Fig. 2.12). Surprisingly, Rho5K321-325S-GFP protein was also found to localize diffusely to the cytosolic compartment (Fig. 2.13). In contrast,

Rsr1K260-264S mutant protein, which carries a similar mutation in the polybasic stretch, localizes to the vacuolar membrane, but no longer localizes to the plasma membrane

(Park et al., 2002). Expression of Rho5C328S and Rho5K321-325S mutant proteins in the rho5Δ strain did not complement the Rho5WT function in mediating cell death in response to

H2O2 (Fig. 2.14). These data suggest that Rho5 requires both the CaaX and the polybasic motif for membrane anchoring and also that the membrane association is necessary for its function. However, these results do not distinguish whether localization to the PM or to

the vacuolar membrane, or to both membranes is required for its role in H2O2–induced cell death using these mutants.

To directly address the question, I looked for a way to target Rho5 specifically to the plasma membrane. Studies in other labs have shown that a 62 a.a C-terminal domain of human RitC GTPase is sufficient to target it to the plasma membrane. Ras1 protein from S. pombe, when fused to this domain, was shown to be specifically targeted to the plasma membrane (Onken et al. 2006). To target Rho5 to the plasma membrane, a

RHO5-RITC-GFP fusion construct was made (see Materials and Methods) and expressed from the RHO5 endogenous promoter in the rho5Δ strain. The Rho5-RitC-GFP fusion construct expressed Rho5 that lacked the polybasic motif and the CaaX motif, which was replaced with the 61 a.a. C-terminal domain of RitC GTPase. When the localization of 36 Rho5-RitC-GFP fusion protein was examined, I found that the Rho5-RitC-GFP fusion protein localized to the plasma membrane (Fig. 2.15, a) with no visible vacuolar membrane localization. However, a small fraction of this protein also localized diffusely to the cytosol of almost all the cells. In addition to this, the protein localized to the

nucleus of many cells (35-40%). Upon H2O2 treatment, Rho5-RitC-GFP exhibited

G12V punctate patterns in the cytosol similarly to the Rho5 -GFP localization after H2O2 treatment. Some of these patches overlapped with the patches from FM4-64 staining (Fig.

2.15, b and data not shown). These data suggest that the Rho5-RitC fusion protein localizes to some endocytic vesicular compartment that costains with FM4-64 after

exposure to H2O2. Further studies are necessary to address the significance of this localization.

Since Rho5-RitC-GFP does not localize to the vacuolar membrane, I asked

whether this protein complemented the WT protein function in mediating H2O2-induced cell death. The rho5Δ strain expressing the Rho5-RitC-GFP fusion protein exhibited a

hypersensitive growth phenotype upon exposure to H2O2 (Fig. 2.16). The sensitive growth phenotype of this strain was much more severe than that of the rho5G12V mutant cells. To further confirm this observation I tested the cell viability of WT, rho5 mutants as

well as the RHO5-RitC-GFP strain after exposing to 4.4mM H2O2 for 1, 2 and 4hr time period (Fig. 2.17). The strain expressing the Rho5-RitC-GFP fusion protein exhibited

G12V very low cell viability after exposure to H2O2 for 1hr compared to the rho5 mutant cells. Cells expressing the Rho5-RitC-GFP protein exhibited cell viability close to 0%

when they were exposed to H2O2 for 2hr (Fig. 2.17). These results suggest that the vacuolar membrane localization of Rho5 is not necessary to mediate cell death after 37 exposure to H2O2. This observation also raise a number of questions including why targeting Rho5 specifically to the plasma membrane makes cells hypersensitive to H2O2

(see Discussion and Future Directions).

The constitutively active rho5G12V exhibits vesicular trafficking defects and fails to

recover from the vacuolar damage caused by H2O2 treatment.

The vacuole plays a central role in the physiology of a yeast cell. It functions as a major site for protein turnover and for storage of toxic substances and cations, and is involved in the maintenance of cytosolic ion and pH homeostasis (Jones et al., 1997).

Glutathione conjugates and complexes of heavy metals and xenobiotics are also transported into the vacuoles, and are sequestered and/or further metabolized (Grant,

2001). I have identified Trr1 (thioredoxin reductase) as a potential downstream target of

Rho5 (see chapter 3). Since Rho5 interacted with Trr1 on the vacuole and other endomembranes (see chapter 3), I investigated whether the vacuoles in WT and rho5 mutants were affected by oxidative stress.

To test the vacuolar function of yeast cells, a thiol-reactive dye CellTracker Blue

CMAC (CMAC hereafter), which freely diffuses across the membranes of live cells was used. Once inside the cell, CMAC undergoes a reaction, which is believed to be GST- mediated, and produces membrane-impermeable glutathione-fluorescent dye adducts

(Kang et al., 2004). CMAC mainly stains the vacuolar lumen in yeast, most likely due to high concentrations of glutathione in the vacuole. Almost all WT and rho5 mutant cells

showed blue fluorescence inside the vacuolar lumen prior to exposure to H2O2, as expected, although the rho5G12V mutant often showed more than three smaller vacuoles

38 per cell (Fig. 2.18). When these cells were stained with CMAC immediately after

removal of H2O2, the staining pattern was significantly different: most of the WT and some rho5Δ cells exhibited the CMAC signal in the cytoplasm rather than the vacuolar lumen (Fig. 2.19, a & b), suggesting that the vacuole became defective. Even though some cells retained CMAC in the vacuolar lumen, their vacuoles appeared as multiple small globes rather than a few (including the central vacuole) per cell. The defect was most severe in the rho5G12V mutant cells. The blue dye was rarely retained in the lumen of

the vacuole in these cells (Fig. 2.19, c), suggesting that H2O2 treatment caused a defect in vacuole function (such as uptaking glutathione conjugates) or in vacuolar membrane integrity.

Next, I asked whether cells could recover from such damage by growing them in

media without H2O2 for an additional 8 hrs following H2O2 treatment, and then staining with CMAC. The rho5Δ cells recovered normal staining of the vacuolar lumen almost completely (Fig. 2.19, e), whereas less than half of the WT cells recovered normal staining and some of them exhibited severely fragmented vacuole morphology (Fig. 2.19, d and 2.22). The rho5G12V mutant exhibited strikingly different CMAC staining: less than

10% of cells recovered the vacuolar lumen staining and even in those cells that retained

CMAC in the lumen, the vacuole often appeared fragmented (Fig. 2.19, f and 2.22).

I also examined the vacuole with FM4-64, which allows the visual detection of vacuolar morphology and bulk membrane traffic, prior to (Fig. 2.20), immediately after

removal of H2O2 (Fig. 2.21 a-c), and after growing for an additional 8hrs without H2O2

(Fig. 2.21 d-f). Although FM4-64 reached the vacuolar membrane, fragmented vacuoles

were observed in the WT and rho5Δ cells soon after removal of H2O2 (Fig. 2.21, a & b). 39 In some WT cells, the fluorescent signal appeared on the plasma membrane or in several patches, suggesting that membrane trafficking from the plasma membrane slowed down

after H2O2 treatment. In addition, small vacuole-like membrane structures or vesicles, which mimic autophagic vesicles (Abeliovich and Klionsky, 2001), appeared within the central vacuole (see arrow, Fig. 2.21, a). Although some cells recovered normal vacuole

morphology after growing without H2O2, about 40% of cells showed fragmented vacuoles

(Fig. 2.21, d and 2.23). The rho5G12V mutant cells showed little staining of FM4-64 on the

vacuolar membrane after H2O2 treatment; instead several patches, which were likely to localize to the early endosomes, were visible near the plasma membrane (Fig. 2.21, c and

2.23), suggesting that cells expressing the constitutively GTP-bound Rho5 were severely

defective in membrane trafficking after H2O2 treatment. Even after growing for 8 hrs

G12V without H2O2, the rho5 cells failed to recover, unlike the rho5Δ mutant (compare e & f in Fig. 2.21). Taken together, these results suggest that the vacuolar function as well as

vesicular trafficking processes become defective upon H2O2 treatment, and Rho5-GTP promotes such damage and/or prevents recovery from such damage.

I hypothesized that the damage to the vacuoles can result in the release of Fe++

(iron) ions and vacuolar proteases into the cytosol. While the presence of excessive Fe++ in the cytosol can cause excess accumulation of ROS, the released proteases can also cause further cellular damage, ultimately resulting in cell death. Alternatively, I cannot rule out the possibility that these vacuolar defects are the end result of cell death (see

Discussion and Future Directions). To distinguish these possibilities, I tested if the cell death phenotype could be rescued by treating the cells with an iron chelator (that will render all the Fe++ unavailable to the cell and prevent subsequent ROS accumulation), or 40 with a protease inhibitor (that will inhibit the vacuolar protease activity and as a result will prevent any cellular damage). I used BPS (bathophenanthroline disulfonate), a lysosomotrophic iron chelator (that has been shown to render iron unavailable to yeast cells) and PMSF (phenylmethanesulphonylfluoride), a protease inhibitor, which has been shown to result in the accumulation of autophagic bodies inside the vacuoles due to impaired protease activity (Takeshige et al., 1992). The growth phenotype of WT and rho5G12V mutant cells was tested after treatment with BPS or PMSF prior to and after

WT H2O2 treatment. Neither BPS nor PMSF could rescue the cell death phenotype of RHO5 and rho5G12V mutant cells (data not shown).

To address whether the cells were dead at the time point when they exhibited vacuolar function or vesicular trafficking defect I also determined the cell viability of WT and rho5 mutant cells by counting the number of methylene blue positive cells, which stains the dead yeast cell blue. I found a very small percentage of methylene blue positive cells in both WT and rho5 mutant cells, suggesting that most of the cells were not dead when defective CMAC staining was observed (Fig. 2.24). Therefore, these observations are consistent with the idea that the vacuolar damage is an early event leading to cell death. However, since BPS and PMSF did not rescue the cell death phenotype of RHO5WT and rho5G12V mutant cells, whether the vacuolar defect and vesicular trafficking defect contribute to cell death or have resulted from cell death still remains unclear (See

Discussion and Future Directions).

41 SUMMARY

This study has uncovered a new role for Rho5 in oxidative stress response in yeast. Presence of a functional RHO5 gene confers sensitivity to oxidants such as DEM,

paraquat and H2O 2. Interestingly Rho5 was found to localize to the plasma membrane and the vacuolar membrane of the cells, suggesting that this protein functions most probably from these subcellular compartments. This observation correlated well with the vacuolar function defect (based on abnormal CMAC staining pattern), and the vesicular trafficking defect (based on the defective uptake of the FM4-64 dye) exhibited by

G12V rho5 mutant cells that express the Rho5-GTP locked form of the protein (after H2O2 treatment). However, to what extent these defects contribute in the cell death process, or whether these defects are an end result of the cell death process is still not clear (see

Discussion and Future Directions). Additionally, vacuolar membrane localization of

Rho5 may not be necessary to mediate cell death after exposure to H2O2. These data suggest that Rho5 could function through some of its downstream target most probably at

the plasma membrane to mediate H2O2-induced cell death process.

42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 CHAPTER 3

Trr1 IS A POTENTIAL TARGET FOR Rho5.

INTRODUCTION

Changes in the intracellular redox state regulate several critical intracellular pathways in mammalian cells (Kamata and Hirata, 1999). All aerobic systems are constantly exposed to these free radicals, and ROS (reactive oxygen species) -mediated cell injury forms the common basis behind a number of disease processes like atherosclerosis, neurological disorders and cancer. In S. cerevisiae, as in higher eukaryotes, ROS are produced as normal byproducts of cellular metabolism. Increased production of ROS triggers defense mechanisms to avoid the deleterious consequence of

ROS accumulation. The glutathione (GSH)/glutaredoxin (GRX) and thioredoxin (TRX) systems are major cellular factors that regulate redox homeostasis by neutralizing free radicals. Thiol groups (-SH) of these components play a remarkably broad range of roles in the cell, and the redox status of the cysteine residues can affect both the structure and function of numerous proteins (Trotter and Grant, 2003) such as reduction of dehydroascorbate, repair of oxidatively damaged proteins, and sulphur metabolism

(Holgrem, 1989; Rietsch and Beckwith, 1998).

In budding yeast, TRR1 encodes the cytoplasmic thioredoxin reductase (Trr1) that reduces the oxidized disulfide form of TRX using NADPH and is required for the

66 protection of yeast cells against oxidative and reductive stress (Trotter and Grant, 2002).

Budding yeast also contains a mitochondrial thioredoxin system, which is thought to protect cells against the oxidative stress generated during respiratory metabolism

(Pedrajas et al., 2000). Trr1 is a known cytoplasmic protein (Chae et al., 1994, Pedrajas et al., 1999, Trotter and grant, 2005) and its level is induced under mild oxidative stress

(for example, upon exposure to 0.3 - 1 mM H2 O2 ) (Morgan et al., 1997, Gasch et al.,

2000).

In chapter 2, I demonstrated that Rho5 is involved in the H2O2-induced cell death process. To understand how Rho5 is involved in this process, I focused on Trr1 because it

co-purified with Rho5 in a high throughput interaction study (Ho et al., 2002) and

because it is a key component in the thioredoxin antioxidant system, which activates

thioredoxin peroxidase and is known to inhibit apoptosis (Kern and Kehrer, 2005). 

In this chapter, I present some data indicating that Rho5 interacts with Trr1 and affects Trr1 levels. These data support the idea that Rho5-GTP contributes to ROS

accumulation and ultimately to cell death upon H2O2 treatment, most probably by lowering Trr1 levels in the cytoplasm.

MATERIALS AND METHODS

Standard methods of yeast genetics, recombinant DNA manipulation, and growth conditions were used unless indicated otherwise (Ausubel et al., 1999; Guthrie and Fink,

1991).

67 Yeast strains used in this study

Yeast strains used in this study are listed in the Strain Table (Table 1).

All strains used in this study were isogenic to HPY210 except as indicated in the Strain

Table.

Staining and Microscopy

Localization of mCherry and GFP fusion proteins was determined using TRITC and FITC filters, respectively, as previously described (Kang et al., 2001), except where

indicated, cells were treated with 4.4 mM H2O2 for 0.5 hr, 1 hr, 2 hr or 4 hr prior to observing under the microscope. Yellow fluorescent signal was visualized using the YFP filter. Image acquisition was carried out using a Nikon E800 microscope (Nikon, Tokyo,

Japan) fitted with a 100X oil-immersion objective (N.A.= 1.30) as previously described

(Kang et al., 2001), except that Slidebook 4 digital microscopy software (Intelligent

Imaging Innovations Inc.) was used to capture a series of optical sections at 0.3-µ intervals. A representative single section from a Z-series of images was used to study the localization pattern of the proteins. The fluorescence intensity of a tagged protein was determined by measuring the fluorescence along the line drawn across the cell and was shown as an intensity plot, where the x-axis represents the length of the line and y-axis represents the fluorescent signal intensity (in arbitrary units).

Cloning, deletion, and tagging of TRR1  

Plasmids and primers used in this study are listed in the Plasmid Table (Table 2) and

Primer Table (Table 3), respectively.

68 To clone the T RR1 gene, PCR was performed using genomic DNA from HPY210

as template and primers oTTR17 and oTRR18. The 2.05-kb PCR product (containing the 

TRR1 ORF plus 592-bp of upstream and 560-bp of downstream sequence) was digested

with S peI (sites included in the primers) and cloned into the SpeI site of pHP1477 (a

pRS426 derivative that  lacks the B amHI site), yielding plasmid pHP1465.  The T RR1

sequence was verified by DNA sequencing.  To construct a TRR1 deletion plasmid,

TRR1 flanking sequence excluding the TRR1 ORF was amplified by PCR using pHP1465

as template and primers oTRR19 and oTRR110. The resulting PCR product is identical to

pHP1465 but lacks the TRR1 ORF. The PCR product was digested with B amHI (sites

included in the primers) and ligated with a 1.7-kb B amHI fragment carrying the HIS3

marker from plasmid pCA5015 (Kang et al., 2004), yielding plasmid pHP1466. The

chromosomal T RR1 gene was deleted in YEF473 (Bi and Pringle, 1996) by one-step gene

disruption (Rothstein, 1991) using the 2.97-kb S peI fragment (carrying t rr1Δ::HIS3) from

pHP1466, yielding HPY1176.  Deletion of T RR1 was verified by genomic PCR, and

meiotic progeny carrying t rr1Δ::HIS3 (HPY1175) were also confirmed by their

sensitivity to oxidants such as diethyl maleate, paraquat, and diamide.

To construct a strain expressing Trr1-Myc, 13Myc-kanMX6 was amplified by

PCR using pFA6a-13Myc-kanMX6 (Longtine et al., 1998) as template and primers oTRR11 and oTRR12. The resulting PCR product was used to transform strain HPY16

(Park et al., 1993), yielding HPY910.

DNA fragment encoding a modified Ds-Red called mCherry (Shaner et al., 2004) was amplified by PCR using pYX1 (kindly provided by Dr. Berl Oakley, The Ohio State

University) as template and primer pair of ORED1CHERRY and ORED2CHERRY. The 69 760-bp PCR product was digested with NotI (sites included in the primers) and cloned into NotI site of pBluescript, yielding pHP1463.

To construct TRR1-mCherry, NotI site was introduced just before the stop codon of TRR1 as follows: The 1.5-kb DNA fragment covering the region upstream of the stop codon was amplified by PCR using pHP1465 as template and primers oTRR17 and oTRR114. In addition, a 560-bp fragment covering the region downstream from the stop codon was amplified by PCR using primers oTRR113 and oTRR18. The 1.5-kb and 560- bp fragments were digested with SpeI and NotI (sites included in the primers) and cloned into the SpeI site of pHP1476, yielding pHP1492. For tagging with mCherry, the 760- base pair NotI fragment encoding Cherry was isolated from pHP1463 and was inserted into the NotI site of pHP1492, yielding pHP1512. The correct orientation of the mCherry insert in pHP1512 was confirmed by digestion with NcoI. To construct TRR1-mCherry in pRS304, pHP1512 was digested with XmaI and SacI, resulting in a 2.7-kb TRR1- mCherry fragment and a larger vector fragment. The 2.7-kb fragment was cloned into the

XmaI-SacI site of pHP1517 and pRS314, yielding pHP1494 and pHP1570, respectively.

HPY210, HPY720, and HPY999 were stably transformed with pHP1494 digested with

SalI, yielding HPY1160, HPY1159, and HPY1161, respectively. To express Trr1 with

triple haemagglutin epitope HA3, the 111-base pair NotI HA3 fragment from pHP835

(Kang et al., 2004) was used to replace mCherry in pHP1570, yielding pHP1496.

70 Yeast two-hybrid assay

A yeast two-hybrid assay was carried out as described by Gyuris et al. (1993). To express a DNA-binding domain fusion to Rho5, PCR was carried out using pHP1280 as template and primers oRHO59 and oRHO510, which also introduced the C328S substitution in Rho5. This mutation was introduced to increase targeting of Rho5 into the nucleus by blocking C-terminal prenylation. The 996-bp PCR product was digested with

BamHI and XhoI, and cloned into the BamHI-XhoI sites of pEG202 (Gyuris et al., 1993), yielding pHP1382. Similarly, DNA fragments carrying rho5G12V and rho5K16N were amplified using pHP1281 and pHP1282, and cloned into pEG202, yielding pHP1386 and pHP1387, respectively. To express an activation domain fusion to Trr1, TRR1 was amplified by PCR using genomic DNA from HPY210 as template and primers oTRR15 and oTRR16. The 960-bp PCR product was digested with EcoRI and XhoI, and cloned into the EcoRI-XhoI sites of pJG4-5 (Gyuris et al., 1993), yielding pHP1383. The yeast strain EGY48 carrying the LEU2 reporter gene (Gyuris et al., 1993) was transformed with pHP1383 in combination with pHP1382, pHP1386, pHP1387, or empty vector, and several independent transformants from each transformation were replica-plated on SC plates lacking leucine (and tryptophan and histidine to maintain the plasmids) to test

Rho5-Trr1 interaction.

Plasmids and yeast strains used for BiFC (Bimolecular Fluorescence complementation)

To construct fusions to the N-terminal fragment of YFP (YFP-N), first, a NotI site was introduced just before the stop codon of TRR1 as follows: The 1.5-kb DNA fragment

71 covering the region upstream of the stop codon was amplified by PCR using pHP1465 as template and primers oTRR117 and oTRR114. In addition, a 560-bp fragment covering the region downstream from the stop codon was amplified by PCR using primers oTRR113 and oTRR18. The 1.5-kb and 560-bp fragments were digested with SpeI and

NotI (sites included in the primers) and cloned into the SpeI site of pHP1476, yielding pHP1492. Second, YFP-N was amplified by PCR using pYN-1 (kindly provided by D.

Bisaro, The Ohio State University) as template and primers oYFP1 and oYFP2. The 477- bp PCR product was digested with NotI and cloned into the NotI site of pHP1492 in correct orientation, yielding pHP1513. To construct pRS304 carrying TRR1-YFP-N, the

2.5-kb XmaI-SacI fragment (TRR1YFP-N) from pHP1513 was cloned into the XmaI-SacI site of pHP1517 (a derivative of pRS304 that lacks the XhoI and SalI sites), yielding pHP1514. The C-terminal fragment of YFP (YFP-C) was amplified using pYC-1 as template and primers oYFP3 and oYFP4. The 252-bp PCR product was digested with

NotI and cloned into the NotI site of pHP1544, pHP1502, and pHP1507 in the correct orientation, yielding pHP1500, pHP1504, and pHP1509, respectively. The correct orientation of the YFP-C insert in these plasmids was confirmed by digestion with BglII.

To construct RHO5-YFP-C in pRS306, the 2.2-kb XmaI-XhoI fragment (RHO5-YFP-C)

from pHP1500 was cloned into the X maI-XhoI site of pRS306, yielding pHP1501. 

Integrating plasmids pHP1506 and pHP1511 were  constructed similarly to express

Rho5G12V-YFP-C and Rho5K16N-YFP-C, respectively. To express TRR1-YFP-N and

RHO5-YFP-C (or Rho5 mutant protein fused to YFP-C) from the chromosomal  TRR1

and RHO5 loci, a r ho5Δ strain (HPY720) was transformed sequentially with pHP1514

after digestion with S alI, yielding HPY1162, and then plasmids pHP1501, pHP1506, or 72 pHP1511, after  S phI digestion, yielding HPY1163, HPY1164, or HPY1165,

respectively. 

Bimolecular fluorescence complementation (BiFC)

The YFP signal was detected using a Nikon E800 microscope as described above except that a single optical section was captured using the YFP filter and by exposing cells to light from mercury lamp for 10 s. Expression of the YFPC-fusion proteins was detected by immunoblotting using Full-length A.v. GFP polyclonal antibody (Clontech

Laboratories, Inc.; Fig. 3.2E).

Protein Analyses

To analyze Trr1 protein, Trr1-HA fusion protein was expressed in WT and rho5 mutants from a low-copy number plasmid, pHP1496. Except as noted, yeast cells were grown to mid-log phase in SC medium lacking nutrients as needed to maintain the plasmid. WT and rho5 mutant cells transformed with pHP1496 were treated with 4.4 mM

H2O2 for 4 hrs. Equal number of oxidant-treated and mock-treated cells were harvested based on the OD600 reading, and the soluble fraction was prepared using a lysis buffer with 0.3% Tween-20 as previously described (Kang et al., 2004). The protein extract was separated on a 10% SDS (sodium dodecyl sulphate) polyacrylamide [made from

30% acrylamide/methylene bisacrylamide solution (37.5:1 ratio)] gel and transferred to nitrocellulose membrane. The HA epitope–tagged Trr1 protein was detected using monoclonal anti-HA antibody HA11 (Covance Research Products, Denver, PA). Protein bands were then detected using fluorescently-labeled secondary antibody

73 IRDyeTM800CW (LI-COR Biosciences, Lincoln, NE) and the Odyssey Infrared Imaging

System (LI-COR Biosciences, Lincoln, NE).



Northern Blotting

WT and rho5 mutants grown to mid-log phase in YPD medium were treated with

3mM H2O2 for 200 minutes. Equal number of oxidant-treated and mock-treated cells were harvested based on the OD600 reading. The cells were suspended in RNA buffer (50 mM Tris (HCl), pH 7.4; 100 mM NaCl; 10 mM EDTA) containing 1.3% SDS, and lysed using glass beads. Total RNA was extracted from the cell extract using equilibrated phenol (pH 7.5) (Sigma), followed by phenol and chloroform extraction, and then chloroform extraction. Total RNA from the acqueous phase was precipitated by incubating the acqueous phase with 4M sodium chloride and 100% ethanol at -80°C for

30 minutes, followed by centrifugation at 14000 rpm for 10 minutes. The total RNA precipitate was washed with 70% ethanol and resuspended in DEPC

(diethylpyrocarbonate)-treated water.

10µg of RNA was separated on 3% polyacrylamide [made from 30% acrylamide/methylene bisacrylamide solution (37.5:1 ratio)] urea gel, was then transferred to Hybond N+ membrane (Amersham Biosciences). The membrane was UV cross-linked and incubated in prehybridization solution (5 x SSC; 50 % Formamide; 5 x

Denhardt's-solution; 1 % SDS; 100 µg/ml heat-denatured sheared non- homologous DNA

(Salmon sperm DNA)) for 4hrs at 42°C. To make a radio-active probe, 15 pmoles of oligo oTRR115 (TRR1 sequence from + 577 to + 627) or 018s1748 (18s rRNA sequence from +1748 to + 1788) was labeled with radioactive 32P using polynucleotide kinase 74 (New England Biolabs Inc.) and ATPγ32P (PerkinElmer). The labeled oligos were purified using minispinTM G-25 columns (Amersham Biosciences) to remove unbound

ATPγ32P.

After prehybridization, the blot was incubated with hybridization solution (same as prehybridization solution, but with 5% dextransulphate in place of salmon sperm

DNA) carrying the 32P labeled oligo oTRR115, overnight at 42°C. The blot was washed once with 2X SSC at room temperature followed by 2X SSC containing 0.1% SDS at

65°C till the background was low. The blot was then exposed to the phosphor imager

(Typhoon 9410, Amersham Biosciences) to detect the TRR1 mRNA signal. The blot was reprobed with radioactive 18s oligo (o18s1748) to detect 18s rRNA (as an internal control).

Purification of GST-Rho5 and in vitro binding assays

Purification of proteins and i n vitro binding assays were performed as previously 

described (Park et al., 1997, Kozminski et al., 2003), except GST-Rho5 was purified

from E . coli (BL21) carrying plasmid pGEX-HA-R HO5 (a gift from F. Doignon, Institut

de Biochimie et Genetique Cellulaires; Roumanie et al., 2001). Briefly, 1 µg of GST-

Rho5 (or GST) was pre-loaded with GTPγS or GDP, and then incubated with lysates

(S10 fraction) prepared from 25 OD600 units of yeast cells (HPY910) expressing Myc-

tagged Trr1.

75 RESULTS

Rho5 interacts with Trr1 in a GTP-dependent manner, in vitro.

As described in chapter 2, Rho5 is involved in the H2 O2 -induced cell death process. To understand how Rho5 is involved in this process, I wished to know what

Rho5 targets are. Among several genes/ proteins that have genetic or physical interactions with Rho5, I focused on Trr1 because it co-purified with Rho5 in a high throughput interaction study (Ho et al., 2002), and Trr1 is an important component of the

Thioredoxin anitoxidant system. To determine whether Rho5 indeed interacts with Trr1

and, if so, whether the interaction is dependent on the GTP-bound state of Rho5, Dr.

Kang carried out an in vitro binding assay, using Rho5 purified as a fusion to glutathione

S-transferase (GST) from bacteria. GST-Rho5, pre-loaded with GTPγS or GDP or in the

nucleotide-empty state, was incubated with yeast protein extract carrying the Myc

epitope-tagged Trr1. When GST-Rho5 was pulled down with glutathione Sepharose

beads from the reaction mixtures, Trr1 co-precipitated with Rho5-GTPγS, but much less

efficiently with Rho5-GDP, nucleotide-empty state of Rho5 or GST control (Fig.  3.1),

suggesting that Trr1 associates preferentially with Rho5-GTP in vitro. 

Rho5 interacts with Trr1 in a GTP-dependent manner, in vivo.

To determine whether Rho5 and Trr1 interact in vivo,  a yeast two-hybrid assay was performed using the L EU2 reporter and plasmids  expressing Trr1 fused to the acidic

activation domain and Rho5 fused to the DNA-binding domain (Fig. 3.2) (Gyuris et al.,

1993). Interaction between the two proteins used in the assay will facilitate yeast cells growth on a medium lackng leucine. Cells expressing Rho5G12V and Rho5, to a lesser 76 extent, grew on the plate lacking leucine. In contrast, cells expressing Rho5K16N (a mutant form of Rho5 that mimics either GDP-locked or nucleotide empty state of the protein) or carrying vector controls grew poorly on the plate lacking leucine(Fig. 3.3, b). The results thus suggest that Trr1 interacts with Rho5-GTP in vivo. When these cells were grown on the DEM containing (and lacking leucine) plates, cells expressing Rho5 grew almost equally well as cells expressing Rho5G12V (Fig. 3.3, c), suggesting that more Rho5 is activated in the presence of DEM.

To monitor whether and where Rho5 interacted with Trr1 in live cells, I employed a bimolecular fluorescence complementation assay (BiFC) (Hu et al., 2002). This technique enables the visualization of protein-protein interactions in vivo. This experiment involves the fusion of the two nonfluorescent halves of yellow fluorescent protein (YFP) separately to two potential interacting proteins. The interaction of the two fusion proteins will result in the reconstitution of the fluorescence (Fig. 3.4) (Hu et al.,

2002).

Trr1 was expressed as a fusion to the N-terminal half of the yellow fluorescent protein (YFP N), while WT and mutant Rho5 proteins were expressed as fusions to the

C-terminal half of YFP (YFP C). Although each truncated YFP fragment was non- fluorescent (data not shown), in vivo interaction between Rho5 and Trr1 brought these

two fragments together, producing a fluorescent signal. Prior to exposure to H2O2, Rho5-

YFPC or Rho5G12V-YFPC, with Trr1-YFPN, exhibited the YFP signal on the vacuolar membrane, whereas little YFP signal was detected with Rho5K16N-YFPC (Fig. 3.5, a-c), suggesting that Rho5-GTP interacted with Trr1 on the vacuolar membrane. The absence of the fluorescent signal in cells expressing Rho5K16N-YFPC and Trr1-YFPN is unlikely

77 to be due to low levels of expression or instability of the mutant protein, since all YFPC fusions were present at about equal levels (Fig. 3.6).

After H2O2 treatment, Rho5-YFPC and Trr1-YFPN co-localized either to the vacuolar lumen of many cells or to the vacuolar membrane, and also often in a patch, which appeared on the vacuolar membrane (Fig. 3.5, d), while Rho5G12V-YFPC and Trr1-

YFPN co-localized in a punctate pattern in the cytoplasm (Fig. 3.5, e). This localization

G12V G12V pattern was similar to that of Rho5 -GFP (after H2O2 treatment), where the Rho5 -

GFP punctate structures were identified as endocytic (and/or secretory) vesicles, based on FM4-64 staining (see chapter 2). Taken together, these data suggest that Rho5-GTP interacts with Trr1 and the interaction occurs at distinct sub-cellular compartments.

Rho5-GTP may inhibit the elevation of Trr1 level upon H2O2 treatment.   

Studies in other labs have shown that Trr1 level is induced under mild oxidative

stress (for example, upon exposure to 0.3 - 1 mM H2O2 ) (Morgan et al., 1997, Gasch et al., 2000). I wanted to know whether Trr1 level was also elevated under the condition

that induced apoptotic cell death and whether the Trr1 localization changed after H2 O2 

treatment in a Rho5-dependent manner. Trr1 was fused to mCherry, a red fluorescent

protein (Shu et al., 2006), and was expressed from its chromosomal locus.  Although the

Trr1-mCherry fluorescent signal appeared throughout the cytoplasm before and after

H2 O2 treatment in WT and r ho5 mutants (Fig.  3.7), the fluorescence intensity was

different in these strains, in particular upon H2O2 treatment. The fluorescence intensities

of over 100 cells of each strain were analyzed by the line intensity plot. The Trr1-

mCherry fluorescence increased about two fold in the r ho5Δ cells at 4 hr after H2 O2 78 treatment (Fig. 3.7). However, there was almost no  increase of the fluorescence in the

rho5G12V cells and very little increase in W T (Fig. 3.7). Trr1-mCherry fluorescence

intensity was also monitored 0.5 h, 1 h, 2 h, and 4 h, after exposure to H2O2  (4.4 mM)

(Fig. 3.8). Consistent with the above data, Trr1-mCherry fluorescence increased in the

rho5Δ cells, but not in the rho5G12V cells. The increase in the Trr1-mCherry level was

more visible in rho5Δ cells after exposure to H2 O2  for 2h (Fig. 3.8). In contrast, Trr1- mCherry level did not change in rho5G 12V mutant cells at similar time points (Fig. 3.8).

Cells expressing Trr1-mCherry were transformed with a plasmid expressing

Vph1-GFP fusion protein. Vph1 is a subunit of vacuolar ATPase that localizes on the vacuolar membrane and function as proton pumps that acidifies vacuolar lumen. The fluorescence intensity of Vph1-GFP was analyzed as control to normalize the Trr1-

mCherry fluorescence intensity. Prior to H2 O2 treatment, the fluorescence intensity of

Vph-GFP was equal between the WT and rho5 mutant cells (322-328) (Fig. 3.9). The

G12V fluorescence intensity of Vph1-GFP was slightly lowered in H2O2 treated rho5  cells

(307±67), but was slightly elevated in H2 O2 treated W T (338±57) and r ho5Δ mutant strain

(360±36) (Fig. 3.9). The Trr1-mCherry fluorescence intensity of H2 O2 treated WT and rho5 mutant cells was normalized to the fluorescence intensity of Vph1-GFP in WT cells

(prior to H2O2 treatment). Trr1-mCherry fluorescence intensity (normalized to the Vph1-

GFP) was elevated in the rho5Δ mutant cells, but did not change much in rho5G12V mutant cells (Fig. 3.9).

To assess the Trr1 levels in WT and rho5 mutants, cells expressing Trr1-mCherry

were also analyzed by flow cytometry. Consistent with the microscopic observation, the

number of cells with higher fluorescence was increased in the rho5Δ mutant, but not in 79 G12V the r ho5  mutant, after exposure to H2O2  {Fig. 3.10; compare the thick red line (+ 

H2 O2 ) to the thin line with filled blue area below (- H2 O2 )}, as analyzed by flow cytometer. 

The Trr1 protein level was also compared in these strains by immuno-blotting HA

epitope-tagged Trr1 (Fig. 3.11). The amount of Trr1-HA level was elevated in rho5Δ, but

G12V not in the  r ho5  mutant cells, after H2O2 treatment (Fig. 3.11). The amount of Trr1

G12V present in r ho5  cells was lower than W T  even prior to exposure to H2O2  (based on

both Trr1-mCherry fluorescence and protein blot). Taken together, these data indicate

that the cytoplasmic Trr1 level is elevated in the r ho5Δ cells, which exhibit little cell

G12V death after H2O2 treatment, but not in the rho5 cells, which undergoes apoptotic cell death. How might Rho5 regulate the Trr1 level? It is possible that Rho5-GTP may be inhibiting the expression of TRR1 or promoting degradation of Trr1 (See Discussion and

Future Directions).

Given the physical interaction between Rho5-GTP and Trr1, and the

accumulation of Rho5-Trr1 complex inside the vacuolar lumen after H2O2 treatment (this study), I favored the idea that Rho5-GTP may be directly involved in regulating Trr1 function. If Rho5 is targeting Trr1 to the vacuole for degradation (I speculated this based

on the BiFC data showing Rho5-Trr1 complex in the vacuolar lumen after H2O2 treatment), I expected to see Trr1 protein accumulate in the vacuolar lumen of the strains lacking vacuolar proteases (due to the lack of enzymes involved in protein degradation). I measured the fluorescence intensity of Trr1-mCherry levels inside the lumen of the

vacuoles of the vacuolar proteases deficient cells (HPY16), prior to and after H2O2 treatment. However, no significant accumulation of the Trr1-mCherry signal was found 80 in H2O2 treated protease deficient cells (data not shown), suggesting that the Trr1- mCherry may not be degraded inside the vacuole. The proteasome-mediated Trr1 degradation still remains as a possibility.

Rho5-GTP may inhibit TRR1 gene expression or affect TRR1 mRNA stability upon

H2O2 treatment.

Northern analysis was used to evaluate the possibility that Rho5 influences the gene expression and TRR1 mRNA stability of TRR1 during normal growth and/or during oxidative stress. I determined the TRR1 mRNA levels in WT and rho5 mutants (prior to

and after treatment with 4.4mM H2O2 for 2hrs and 4hrs) by northern blotting. While I could detect a significant decrease in the TRR1 mRNA level in rho5G12V mutant cells and

WT cells in these conditions, I could also detect a slight reduction in the 18s rRNA

(internal control) level in all the oxidant-treated samples, when compared with the control samples (data not shown).

Because it has been reported that 18s rRNA does not undergo any significant

degradation after treatment with 3mM H2O2 for 200 minutes (Mroczek and Kufel, 2008),

I tested the TRR1 mRNA levels from cells treated with similar conditions of oxidative stress. I tested the growth phenotype of WT and rho5 mutant cells after treatment with

3mM H2O2 for 200 minutes (Fig. 3.12), and observed a similar pattern of growth sensitive phenotype between the WT and rho5 mutant cells, in the presence of 3mM H2O2 for 200 minutes (Fig. 3.12), as seen after treatment with 4.4mM H2O2 for 4 hrs (Chapter 2, Fig.

2.16).

81 Since the rho5G12V mutant cells exhibited cell death after treatment with 3mM

H2O2 for 200 minutes, total RNA was extracted from WT and rho5 mutant cells after treatment with similar conditions of oxidative stress. TRR1 mRNA level was significantly decreased in rho5G12V mutant cells (1:0.37; ratio of TRR1 mRNA level prior to and after

H2O2 treatment) and WT cells (1:0.45; ratio of TRR1 mRNA level prior to and after H2O2 treatment) after treatment with 3mM H2O2 for 200 minutes (Fig. 3.13). There was little reduction in the TRR1 mRNA levels in rho5Δ cells (1:0.79; ratio of TRR1 mRNA level

prior to and after H2O2 treatment) after H2O2 treatment, providing evidence that Rho5 may regulate (directly or indirectly) the transcript TRR1 level during oxidative stress.

However, the drastic reduction in the TRR1 transcript level during oxidative stress in the rho5G12V mutant cells is in sharp contrast with the relatively minor change in the Trr1

protein level (1:1.01; ratio of Trr1-mCherry fluorescence intensity prior to and after H2O2 treatment). In addition to this, the Trr1 protein level was elevated in rho5Δ cells (1:1.7;

ratio of Trr1-mCherry fluorescence intensity prior to and after H2O2 treatment) after H2O2 treatment contrary to a slight reduction in the TRR1 mRNA level in these cells, indicating that additional post-translational regulatory mechanisms control the Trr1 protein levels inside these cells (see Discussion and Future Directions).

TSA1 overexpression rescues H2O2-induced cell death in WT cells to a small extent

Since lower Trr1 level is likely to result in lower thioredoxin peroxidase activity,

which reduces H2O2 (Chae et al., 1994), we asked whether overexpression of TSA1, which encodes a cytoplasmic thioredoxin peroxidase, suppresses apoptotic cell death of

WT or the rho5G12V mutant. The WT cells carrying the multicopy TSA1 plasmid exhibited 82 higher cell viability (~47%) after H2O2 treatment compared to the vector control (~36%), suggesting that overexpression of TSA1 can partially suppress H2O2-induced cell death

(Fig. 3.14). However, the rho5G12V mutant carrying the same plasmid exhibited little increase of cell viability, suggesting that Tsa1 is not active in cells expressing the GTP- locked Rho5.

SUMMARY

Trr1 was found to interact with Rho5 in vivo, preferably with the Rho5-GTP (i.e. the active state of the protein). Studies on other GTPases suggest that most of the targets of the GTPases bind to the GTP-bound state of the protein. Hence it is very likely that

Trr1 is acting downstream of Rho5, and is a potential downstream target of Rho5. Trr1 is a key component of the thioredoxin antioxidant system, facilitating cell survival during

oxidative stress. Rho5 on the other hand mediates H2O2-induced cell death, suggesting that both these proteins play opposing function in the cells.

Rho5 was also shown to interact with Trr1 on the vacuolar membrane, and also

G12V inside the vacuolar lumen of many cells after H2O2 treatment. Rho5 -Trr1 interaction

G12V appeared in a punctate pattern in the cytosol after H2O2 treatment (similarly to Rho5 -

GFP localization after H2O2 treatment). However, the physiological significance of these interactions in different subcellular compartments is still not clear.

Surprisingly, Rho5 was also found to affect (directly or indirectly) Trr1 levels, both at the protein level as well as at the transcriptional level. These observations raise a number of questions: Does Rho5 target Trr1 for degradation? Does Rho5 affect TRR1 mRNA stability or gene expression? Future studies will be aimed at determining the

83 significance of these interactions in different cellular compartments, and also testing the mechanism by which Rho5 regulates Trr1 (see Discussion and Future Directions).

84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 CHAPTER 4

DISCUSSION AND FUTURE DIRECTIONS

Oxidant-induced cell death has been reported in various cell types including mammalian cells, plants, and yeast (Madeo et al., 1999, Cai, 2005, Kawasaki et al.,

1999). Although small GTPases have been implicated in apoptosis (Kawasaki et al.,

1999, Bivona et al., 2006) and ROS production (Diekmann et al., 1994, Yang, 2002), the mechanisms underlying such cellular responses are poorly understood. This work has

uncovered a new role of the Rho5 GTPase in H2O2-induced cell death in budding yeast.

Rho5 is likely to trigger a signal transduction pathway in response to H2O2 to execute cell death, since cells lacking Rho5 exhibit little cell death after H2O2 treatment. As discussed in Chapter 2, Rho5 mediated cell death is accompanied by the characteristics of apoptosis such as DNA fragmentation and ROS accumulation. By analogy with DNA damage- induced apoptosis (Norbury and Zhivotovsky, 2004), I propose that Rho5-GTP promotes apoptosis-like cell death in yeast cells with excess oxidative damage.

 Despite its sequence similarity to Rac GTPases in mammals, Rho5 is likely to

regulate  ROS accumulation in a distinct way. This is because Rac triggers ROS

production by activating NADPH oxidases in various cell types (Kawasaki et al., 1999,

Diekmann et al., 1994), while no such gene in S. cerevisiae has been directly annotated

as NADPH oxidase. As discussed in Chapter 3, I identified Trr1 as a potential

100 downstream target of Rho5. While the phenotype of rho5 mutants in the oxidative stress conditions and its interaction with an antioxidant component (Trr1) suggest potentially interesting functions of Rho5 in the cells, many questions still remain unanswered as discussed below.

How does Rho5 function in the H2O2-induced cell death process?

Where does Rho5 function in mediating the H2O2-induced cell death process?

Rho5-GFP fusion protein localized to the plasma membrane as well as to the vacuolar membrane of the cells. I wanted to test if Rho5 localization to either one of the

compartments is necessary or sufficient for its function in the H2O2-mediated cell death process. By targeting Rho5 to the plasma membrane of the cells, I expected to see the phenotype either sensitive similar to that of the WT strain, or resistant similar to the

rho5Δ strain to H2O2 treatment.

The Rho5-RitC-GFP fusion protein localized to the plasma membrane of the cells

(not to the vacuolar membrane) and diffusely to the cytosol of most of the cells. In addition to this, a small fraction of this protein localized inside the nucleus of many cells.

Expression of Rho5-RitC-GFP fusion conferred hypersensitivity to H2O2 that was much more severe than that of the rho5G12V mutant. Since the Rho5-RitC mutant protein did not localize to the vacuolar membrane, the vacuolar membrane fraction of Rho5 may not

function in the H2O2-induce cell death process. However, the contribution of the nuclear fraction and the plasma membrane fraction of Rho5 in this process remains unclear.

101 Why does Rho5-RitC fusion protein confer hypersensitivity to H2O2, and what is the physiological significance of punctate pattern of localization after H2O2 treatment? I speculate that Rho5WT protein participates in the vesicular trafficking process, which may be necessary for plasma membrane targeting of the newly synthesized protein and also for normal recycling between the plasma membrane and internal compartment of the cells. Targeting Rho5 to the plasma membrane through the RitC domain or creating an activating mutation in Rho5 (i.e. Rho5G12V), may be causing a severe impairment in the

vesicular trafficking processes after H2O2 treatment. This could explain the punctate localization pattern observed after H2O2 treatment. However, it is still not clear whether the plasma membrane fraction of Rho5-RitC fusion protein, the newly synthesized protein, or both localize to the endocytic vesicles. The C-terminal domain of RitC

GTPase does not have the lipidation motif typically found in Ras family GTPases.

Instead, RitC GTPase has a cluster of basic amino acids that is believed to mediate the membrane association of this protein by a mechanism that is poorly understood (Lee et al., 1996). Rho5WT protein, with a functional CaaX and polybasic motif at its C-terminus may participate in the vesicular trafficking process in a manner different from that of the

Rho5-RitC fusion protein. This could explain the severity of the sensitive phenotype of

the cells expressing Rho5-RitC fusion protein after H2O2 treatment. Despite the fact that expression of Rho5-RitC fusion results in much less cell survival than the WT, the above data suggest that the plasma membrane fraction of Rho5 is more likely to function in the

H2O2-mediated cell death process.

With the existing data, I still cannot confidently conclude that the plasma

membrane fraction of Rho5 localizes to the endocytic vesicles after H2O2 treatment. The 102 possibility of the newly synthesized protein being entrapped in the endosomes still remains. The Rho5G12V-Trr1-YFP complex (that normally localizes on the vacuolar

membrane) also resides in the punctate compartment after H2O2 treatment. All these data suggest the presence of a complex mechanism regulating the localization of Rho5 to different cellular compartments.

Future experiments will be aimed at identifying endocytosis mutants in which the

G12V Rho5 protein fails to localize to the punctate structures in the cytosol (after H2O2 treatment), or the plasma membrane localization of Rho5 is impaired. Experiments determining the cell viability of such endocytosis mutant cells will shed light on the molecular mechanism of Rho5-mediated cell death process. It is also possible that Rho5 interacts with its downstream effector, most probably on the plasma membrane, to activate the cell death process. Future experiments will aim at identifying the regulators and downstream effectors of Rho5 involved in this process. Growth phenotype of yeast genome-wide deletion mutant collection expressing rho5G12V mutant protein in the

presence of H2O2 can be tested. A deletion mutant that does not undergo cell death is most likely lacking the protein functioning downstream of Rho5 in facilitating the H2O2- induced cell death process. In addition to this, a multi-copy suppressor screen can be performed to identify genes when overexpressed can rescue the cell death of WT cells.

Genes whose overexpression suppress the cell-death phenotype of WT cells are likely to have a function opposite to that of Rho5 in the cells.

Do vacuolar function defects and/ or vesicular trafficking defects contribute in Rho5- mediated cell death process?

103  Given the recovery of the vacuoles after removal of H2O2 in the rho5Δ mutant, the vacuolar defects are likely to be the cause, rather than the consequence of cell death.  I speculated that the vacuolar damage might cause the release of metal ions such as Fe++ and vacuolar proteases (that are normally stored in the vacuoles) into the cytosol. As a result, more ROS is produced in the cytoplasm (by the Fenton reaction) (Veal et al.,

2007), contributing to cellular damage, ultimately leading to cell death. Alternatively, the

vacuolar defect could be the consequence of cell death after H2O2 treatment. To distinguish these possibilities, I tested if treatment of cells with an iron chelator or with a

protease inhibitor can rescue the cell death phenotype after H2O2 treatment. Treatment of cells with an iron chelator or with a protease inhibitor did not rescue the cell death phenotype of RHO5WT and rho5G12V mutant cells (data not shown).

Since treatment of cells with BPS or PMSF did not rescue the cell death

G12V phenotype of WT or rho5 mutant cells after H2O2 treatment, one straight-forward interpretation of this result is that these reagents may not function as effectively as documented in the literature, or in the conditions that I have tried. Additionally, I also tested the viability of the cells based on methylene blue staining (this reagent stains dead yeast cells blue). This experiment allowed me to assess the cell viability immediately

after H2O2 treatment and during recovery. Surprisingly, a very small percent of methylene blue positive cells were found despite abnormal CMAC staining. These data suggest that most of the rho5G12V cells are committed to undergo apoptotic cell death upon exposure to

H2O2 but that cell-death execution occurs much later. To summarize the above observations, the methylene blue staining data support the idea that vacuolar damage is an early event in cell death. Although treatment with an iron chelator or a protease 104 inhibitor alone did not rescue the cell death, future experiments combining different iron chelators or combining an iron chelator with a protease inhibitor can be performed to

determine the cell death phenotype after H2O2 treatment. In addition to this the growth phenotype of protease deficient strains after expressing Rho5WT and Rho5 mutant proteins

can be examined after H2O2 treatment, to test the role vacuolar proteases play in mediating the cell death process.

The rho5G12V mutant cells also exhibited an impairment in the uptake of FM4-64 dye (a stryryl dye that enters the cell by active endocytosis and stains the vacuolar membrane) to the vacuolar membrane, suggesting a vesicular trafficking defect. Hence, a membrane trafficking defect could contribute to cell death.

Consistent with this idea, studies in other labs have suggested a direct link between trafficking and oxidative stress induced programmed cell death (PCD) in yeast and arabidopsis cells (Levine et al. 2001, Lin et al. 1996). Studies have also shown that

contrary to what is widely believed, H2O2 does not freely diffuse across biomembranes, and that the regulation of plasma membrane permeability to H2O2 is another mechanism by which cells respond to and adapt to H2O2 (Branco et al. 2004). Given that vesicle trafficking processes form an efficient mechanism to repair damage of the plasma membrane by recycling the membrane components, Rho5 could possibly be interfering

with the cellular trafficking system to facilitate cell death after H2O2 treatment. Future experiments can be performed testing if overexpression of VAMPs (vesicle associated membrane protein are a group of proteins involved in vesicle trafficking in yeast), such as

SNC1 and SNC2 could rescue the cell death phenotype of WT and rho5G12V mutant cells.

105 This experiment will help test to what extent the vesicle trafficking and/or membrane

recycling processes participate in the cell death process mediated by H2O2.

How does Rho5 regulate Trr1?

 Rho5 shares considerable sequence homology to Rac GTPases in mammals,

which is shown to trigger ROS production by activating NADPH  oxidases in various cell

types (Kawasaki et al., 1999, Diekmann et al., 1994). No gene in S. cerevisiae has been

directly annotated  as NADPH oxidase.

As discussed in Chapter 3, Rho5 interacts with Trr1 in a GTP-dependent manner.

Trr1 activates TRXs, which in turn activates thioredoxin peroxidase, thus detoxifying

H2O2 and other reactive  oxygen species (Chae et al., 1994), thereby facilitating cell survival. Since Rho5 is necessary for ROS accumulation and cell death after H2O2

treatment (discussed in Chapter 2), Rho5 and Trr1 may function in an antagonistic

manner under oxidative stress. Given that most of the downstream targets of GTPases

specifically interact with the GTP-bound state of GTPases, I favor the idea that Trr1 is a

downstream target of Rho5 and that Rho5-GTP may inhibit the Trr1 function. However, the recombinant Rho5 did not inhibit Trr1 activity in vitro (S. W. Kang, Ehwa University,

Korea, personal communication). It is still possible that Rho5-GTP undergoes additional modification in vivo, which is required for its function to inhibit Trr1. This is also reflected in the fact that rho5G12V mutant cell (that expresses Rho5-GTP locked form of

the protein) does not exhibit any growth defect prior to H2O2 treatment when compared with the WT cells. Alternatively, Rho5-GTP may be involved in lowering Trr1 level,

106 since the Trr1 protein level was elevated in the rho5Δ cells after H2O2 treatment, but not

in the r ho5G12V cells.

Very few biochemical studies have been performed to test the protein profile of cells undergoing apoptotic cell death (caspase-dependent or caspase-independent process). Recently, Mahrus et al. (2008) and Dix et al. (2008) have used new approaches to profile proteolytic events in apoptotic Jurkat cells. They identified 240 and 170 new substrates of caspases, respectively, in their studies (with very little overlap). Even though further studies need to be done to verify each putative caspase substrate, these studies suggest the presence of a large-scale (regulated) proteolytic event occurring inside apoptotic cells. Given the scale of proteolysis, it seems more likely that some protein cleavage is important and contributes to cell death, while many other proteins are just caught up in the proteolysis maelstrom accompanying apoptosis and cell death process.

Yeast cells may also face similar large-scale proteolysis during H2O2-induced cell death. Consistent with this idea, I observed that in addition to Trr1, a few other proteins

G12V (α-Tubulin, Bud4, Rna1) level were lowered in the H2O2-treated rho5 mutant cells

(data not shown). Further study is required to test the mechanism by which Rho5 regulates Trr1 protein and mRNA level. Future experiments will also be aimed at test if there are additional substrates undergoing degradation during the Rho5-mediated cell death process.

Since the BiFC data exhibited Rho5-Trr1 interaction inside the vacuolar lumen

after exposure to H2O2, I hypothesized that the translocation of Rho5-Trr1 complex into the vacuolar lumen contributes to the decrease of the Trr1 level in the cytoplasm. In other words, Rho5-GTP may target Trr1 to the vacuole to promote its degradation or to exclude 107 Trr1 from its substrate. However, the fluorescence intensity of Trr1-mCherry signal did

not increase inside the vacuole of the protease deficient cells after H2 O2 treatment, suggesting that Trr1 is not undergoing degradation in the vacuole after H2O2 treatment.

Interestingly, Rho5G12V-Trr1-YFP complex localized to the punctate compartment (most

probably the endocytic compartment) in the cytosol after H2O2 treatment. This raises the

G12V possibility that prior to H2O2 treatment, the Rho5 -Trr1 interaction occurs initially inside the endocytic compartment of the cells, which then translocate to the vacuolar

membrane. Defective vesicular trafficking after exposure to H2O2 treatment may inhibit the translocation of the Rho5G12V-Trr1 complex to the vacuolar membrane. This possibility is further strengthened by studies showing that formation of YFP complex is irreversible in vitro (Hu and Kerppola 2005). Although the above possibility cannot be ruled out, it is very likely that Rho5G12V-Trr1-YFP complex localization to the endocytic compartment is an end result of oxidative stress because the Rho5G12V-GFP fusion protein

localizes in a similar pattern after H2O2 treatment.

I found that the TRR1 transcript level was decreased in the cells expressing Rho5-

GTP locked form of the protein (i.e. rho5G12V mutant cells). Low levels of TRR1 transcript

in the H2 O2 treated cells could possibly be due to either a reduction in the TRR1 gene expression or to reduction in the mRNA stability. Studies in other labs have shown that the human thrioredoxin reductase (TrxR; a selenoenzyme) has in its 3'UTR (3' untranslated region) elements influencing TrxR mRNA stability (Gasdaska et al., 1999).

Hence, a reporter based expression assay using constructs containing TRR1 promoter

sequence, 3' UTR, and/or 5' UTR, can be performed prior to and after H2 O2 treatment, to test the presence of similar elements influencing TRR1 mRNA stability.

108 The different level of TRR1 mRNA level in each strain before and after H2 O2

treatment is not quite reflected in the protein level. For instance, there is a slight

reduction in the TRR1 mRNA level in rho5Δ strain after H2O2 treatment, which is opposite to the slight elevation in the Trr1 protein level observed in this strain under similar condition of stress. These observations favor the presence of additional regulatory mechanisms (at the translation level) affecting Trr1 function. Determining the Trr1

protein stability and turn over rate (prior to and after H2 O2 treatment) inside the cells could shed light on its regulation under these conditions.

Low Trr1 levels in rho5G12V cells may result in low thioredoxin peroxidase (Tsa1)

activity, and thus promote excess ROS accumulation and subsequent cell death. Although

the increase of the Trr1 level in the r ho5Δ mutant after H2O2  treatment does not appear to

be substantial, it might be critical for how efficiently cells can detoxify H2O2, particularly

when cells are exposed to exogenous H2O2. H2O2 can trigger high accumulation of ROS

by iron catalyzed reactions  (called Fenton reaction; Veal et al., 2007). Consistent with

this idea, I found that overexpression of TSA1 partially suppresses the cell death of W T

upon exposure to H2O2. However, overexpression of  T SA1 does not suppress H2O2-

induced cell death of the r ho5G12V cells, suggesting that Tsa1 may not be functional/stable

in cells expressing GTP-locked Rho5 despite its overexpression. However, these genetic data provide only indirect evidence, and alternative interpretations are also possible.

Further molecular analyses would be necessary to understand the mechanism by which

Rho5 regulates the TRX system.

BiFC data (described in Chapter 3) showed that Rho5 interacts with Trr1 on the vacuolar membrane. However, the growth phenotype assay of cells expressing Rho5- 109 RitC fusion protein suggested that the vacuolar membrane fraction of Rho5 may not be

necessary to mediate H2O2-induced cell death. This raises an important question about the role of Rho5-Trr1 interaction in the cell death process. The genetic data showing that overexpression of TSA1 rescues the cell death phenotype supports Trr1 as a possible

downstream target of Rho5 in mediating the cell death process after exposure to H2O2.

However, to what extent the Trr1 interaction with Rho5 contributes to H2O2-induced cell

death is still not clear. There is a possibility that Rho5-Trr1 interaction may be playing a more prominent role in a different type of oxidative stress condition (mediated by a different oxidant). Interestingly, the yeast two-hybrid data indicate that Rho5-Trr1 interaction increases on the DEM-containing plate, suggesting that more Rho5 is converted to the GTP-bound state in the presence of DEM. In addition, the trr1Δ conferred sensitivity to DEM (data not shown). Interestingly, the cell death phenotype of

G12V rho5 mutant is more severe than the trr1Δ after exposure to H2O2.These observations

suggest the presence of additional downstream targets of Rho5 in mediating H2O2-

induced cell death.

Does Rho5 undergo post-translational modification to mediate cell death?

G12V Since rho5 mutant cells did not exhibit cell death prior to H2O2 treatment, the

Rho5-GTP may require further activation upon exposure to H2O2, perhaps through a post- translational modification or interaction with another protein. A recent study in another lab has reported that Rho5 is phosphorylated by Npr1 kinase and this modification seems to be necessary for further modification of Rho5 by ubiquitination and its subsequent degradation (Annan et al., 2008). It will be interesting to test if oxidative stress changes/ 110 affects the phosphorylation state of Rho5 and also test the stability of Rho5 after

exposure to H2O2. Whether Rho5 undergoes proteasome-mediated degradation, and whether this degradation is the primary mechanism by which Rho5 is regulated are some other interesting questions that remain to be answered.

Why have the cells retained the Rho5-mediated cell death program?

The next big question relevant to this work is why have cells retained the RHO5 gene that confers sensitivity under stress conditions. What benefits do cells gain by retaining the Rho5-mediated cell death process? There is a growing evidence to support that yeast cells living in a colony constantly sieve out the older cells to facilitate growth of newer, healthier cells in the presence of limiting nutrients (Buttner et al., 2006). Under these conditions the ability to selectively kill the older, unhealthy cells that forms a fraction of the colony will be beneficial for the overall population of cells, as all the cells are trying to survive on the limiting nutrient source. One idea is that Rho5 could be acting as a sensor to selectively trigger the cell death process in old and irreversibly damaged cells. If this is the case, the presence of Rho5 could be beneficial to cells on a long-term basis, especially in the aging cultures. I tested this hypothesis by comparing the cell viability of WT and rho5Δ strains cocultured for a prolonged period of time (see

APPENDIX B). After the initial period of similar viability, there was a significant reduction in the viability of the rho5Δ strain, while the WT cells continue to grow and eventually populated the entire culture. Control experiment with gic2 deletion cells and

WT cells did not exhibit any difference in the cell survival abilities, suggesting that the presence of Rho5-mediated cell death is responsible for better survival of the WT cells. 111 The above experiment and similar experiments in other labs with other mutant strains, have identified YCA1 (metacaspase) (Madeo et al., 2002) and AIF1 (apoptosis- inducing factor) (Wissing et al., 2004) to be beneficial to yeast in ageing culture, thus supporting the idea of altruistic cell death program in unicellular organisms.

Identification of these and other components that are conserved from yeast through humans, along with understanding their mechanism of function, could have far-reaching implications in evolutionary biology. Future studies investigating the molecular mechanism of Rho5- mediated ROS accumulation and cell death process will hopefully help identify more such players specifically functioning in this pathway.

Yeast and Apoptosis?

Several recent reports suggest that yeast cells undergo "programmed cell death" in response to various signals such as oxidants and acetic acid, and also during mating and aging, but whether all these cases are truly apoptotic (as a caspase-mediated death process) or non-apoptotic cell death is unclear (Hardwick and Cheng 2004, Buttner et al.,

2006). Studies in these labs have also shown that yeast exhibits other characteristics of apoptotic mammalian cells, such as exposure of phosphatidylserine (PS) to the exterior surface, and chromatin condensation. But one should be careful with using the above criteria to support the existence of apoptosis in yeast, as it is still not clear if the event leading to apoptosis -like morphology contribute to yeast cell death (Hardwick and

Cheng, 2004). I did observe DNA fragmentation (by TUNEL staining) in cells undergoing Rho5-mediated cell death, however, I could not observe other hallmarks of apoptosis such as exposure of the phosphatidylserine on the cell surface and exclusion of

112 propidum iodide staining (data not shown). So the question if Rho5-mediated cell death is truly apoptotic (which is a caspase-mediated death process), or it is just an 'apoptotic-like cell death' process exhibited by some simple eukaryotes like budding yeast still remains.

Interestingly, studies in other labs have identified proteins such as Yca1 (Madeo et al., 2002, Khan et al., 2005), Aif1 (Wissing et al., 2004), and Nuc1 (endonuclease G)

(Buttner et al., 2007), which have been implicated in apoptotic cell death in yeast. These observations raised the question whether the Rho5-mediated cell death program involves other proteins such as Yca1, Aif1, and Nuc1. I took a candidate approach to address the question (see APPENDIX C). I concluded from this experiment that none of the proteins reported to be involved in the yeast programmed cell death process acted in concert with

Rho5 to mediate H2O2-induced cell death (see APPENDIX C). Future studies will be necessary to determine whether the Rho5-mediated cell death process defines a distinct cell death program. What other cell death processes exist in yeast and to what extent these cell death programs are conserved between yeast and other organisms are other interesting questions that remain to be answered. Another question that is relevant to this work is the possible benefits a unicellular organism like budding yeast will gain to undergo apoptosis, a process whose benefit in a multicellular system is undisputed.

Further study needs to be done to confirm if the yeast cell death process is truly apoptosis.

113 APPENDIX A

TEST THE ROLE OF Rho1-GTPase IN OXIDATIVE STRESS RESPONSE IN

YEAST.

INTRODUCTION

Rho1 is an essential gene belonging to the Rho family of GTPase and it has been shown to bind to and activate Fks1 and Fks2 - two closely related subunits of 1,3-β- glucan synthase (Mazur and Baginsky, 1996) - thereby controlling cell wall synthesis.

Rho1 has also been shown to bind to and activate Pkc1, a yeast homologue of mammalian Protein Kinase C, which participates in activating Mitogen Activated Protein

Kinase (MAPK) pathway (Lee and Levin, 1992; Kamada et al., 1995). This pathway has been shown to regulate the actin cytoskeleton and the expression of genes involved in cell integrity.

Several conditional lethal mutations (high-temperature-sensitive mutations) of

RHO1 have been isolated in other labs to characterize the function of Rho1 inside the cells (Saka et al., 2001). Several labs have shown the involvement of Rho1 and upstream regulatory components of Rho1 in various stress pathways such as osmotic stress and heat shock stress (Kamada et al., 1995; Harrison et al., 2004). Rho1 has also been shown to be involved in resistance to the heavy-metal toxicity in yeast most probably by regulating the function of ABC transporters residing on the vacuolar membrane (Paumi et

114 al., 2007). Recent studies have also shown the involvement of Rom2 (a GEF for Rho1) in the oxidative stress response in yeast (Vilella et al., 2005). Given the role Rho1 plays in various stress pathways, and that various upstream and downstream elements of Rho1 play in the oxidative stress response, I decided to test the function of Rho1 in oxidative stress response pathway in yeast.

MATERIALS AND METHODS

Growth phenotype in the presence of oxidants

ts WT and rho1 mutants were diluted to OD600 = 0.4 from a mid to late logarithmic phase culture in YPD and then serially diluted (10-fold). These cells were spotted on

YPD plates containing 2mM H2O2, 400µg/ml paraquat, 1mM diethyl maleate or no oxidant. The plates were incubated at 30°C for 2-7 days. To test the growth phenotype of

ts rho1 mutants in the presence of H2O2, cells from the mid logarithmic phase culture were diluted to OD600 = 0.2. 200 µl of the diluted culture was spread on YPD plates to make a lawn of cells. Filter disc (Whatman filter paper) soaked in 5µl of H2O2 (of concentrations

1M and 4M) was placed on the plate. The plates were incubated at 30°C for 1-2 days before measuring the area of no growth.

Some pkc1 mutants (gift from D. Levin's lab, Johns Hopkins University) were also tested in the presence of H2O2. When testing pkc1 mutants, cells from the mid logarithmic phase culture were diluted to OD600 = 1 and then serially diluted (5-fold). 200

µl of these cultures were transferred to 96-well plates in the presence of 1, 2 and 3mM

H2O2 and no oxidant and incubated for 4hrs and plated on SD plates containing all the

115 amino acids and nucleotides except uracil with or without 1 M sorbitol using a sterile metal pins (data not shown).

The growth phenotype of yeast strains lacking some ABC transporter genes

(purchased from the Open Biosystem) and tus1Δ strain (purchased from the Open

Biosystem) were tested in the presence of H2O2 as described above (i.e. on YPD plates containing 2mM H2O2). Yeast strain BY4741 was used as WT control when using yeast deletion mutants purchased from Open Biosystem.

Staining and Microscopy

GFP-Rho1 fusion protein was expressed from a URA3 marked plasmid (pGFP-N-

FUS-RHO1 [Marelli et al., 2004]) under the methionine repressible promoter (pGFP-N-

FUS [Niedenthal et al. 1996; Marelli et al., 2004]). Cells were grown in minimal

(synthetic dextrose) media carrying all the nutrients except uracil (SD-URA) to maintain the plasmid. To express the fusion protein, cells growing in SD-URA media were harvested and resuspended in synthetic dextrose media carrying all the nutrients except uracil and methionine (SD-URA, MET). GFP-Rho1 fusion protein expression was induced for seven hours and the cells were observed under the microscope.

Staining with FM4–64 (Invitrogen) was performed for 40 minutes as described by

Vida and Emr (1995). Localization of GFP-Rho1 was determined as described in Kang et al. (2004) except that, where indicated, cells were treated with 2mM H2O2 (for 4hrs) before they were observed under the microscope. Image acquisition was carried out by using a Nikon E800 microscope fitted with 100x oil-immersion objective (N.A. = 1.30) as described in Kang et al. (2004), except that Slidebook 4 digital microscopy software

116 (Intelligent Imaging Innovations) was used to capture a series of optical sections at 0.3-µ intervals. A representative single section from a Z-series of images was shown to study the localization pattern of the proteins.

RESULTS

Some rho1ts mutants were sensitive to different oxidants.

To test if Rho1 affects the redox balance in yeast, I first examined the growth phenotype of some of the rho1ts mutants (rho1E45V [rho1-2], rho1L60P [rho-3] and rho1G121C [rho1-5]) (Gifts from Erfei Bi' lab, University of Pennsylvania) in the presence

G121C of oxidizing agents such as DEM, paraquat and H2O2. The rho1 (rho1-5) mutant was found to be sensitive to paraquat, DEM and H2O2 (Fig. 4.1 & 4.2) at 30°C (permissive

E45V temperature). Interestingly, rho1 (rho1-2) mutant was sensitive to H2O2, but not to

DEM and paraquat (Fig.4.1 & 4.2). Studies in other labs have shown that both these mutants fall in the same complementation group and they fail to activate Pkc1-MAPK pathway and also possessed delocalized actin patches (Saka et al., 2001). This led me to think that Rho1 could possibly be signaling via the Pkc1-MAPK pathway during oxidative stress response. Consistent with this idea, some recent work from another lab reported that Pkc1 and some MAPK components are involved in cellular response to oxidants such as Diamide and H2O2 (Vilella et al., 2005).

I then tested the growth phenotype of pkc1 mutants (gift from D. Levin's lab

Johns Hopkins University) (Levin et al., 1990). Specifically, the growth phenotype of pkc1Δ strain carrying plasmid expressing pkc1-1, pkc1-2 and pkc1-3 mutant protein were tested in the presence of H2O2. Surprisingly little growth differences were found among 117 the cells expressing wild-type PKC1 and various pkc1 mutant genes (data not shown).

There is a good possibility that the sensitivity of pkc1 mutant is dependent on the strain background. Rho1 may still function in oxidative stress by using other components of the

Pkc1-MAPK pathway, but independently of Pkc1.

To test whether any other downstream components of Pkc1-MAPK pathway in involved in the oxidative stress response, I tested the growth phenotype of bck1Δ and mpk1Δ in the presence of H2O2. I found that mpk1Δ conferred mild sensitivity to H2O2

(Fig. 4.3).

Localization of GFP-Rho1 prior to and after H2O2 treatment.

To further understand the role of Rho1 in the oxidative stress response process, the localization of GFP-Rho1 fusion protein prior to and after H2O2 treatment was studied. Previously, work by immunofluorescence and live-cell imaging have shown that

Rho1 localizes to the site of bud emergence, the tip of the growing bud and mother-bud neck region before cytokinesis (Yamochi et al., 1994) (Fig. 4.4). GFP-Rho1 exhibited a similar localization pattern in these cells prior to H2O2 treatment (Fig. 4.4). In contrast, after H2O2 treatment, GFP-Rho1 enriched on the vacuolar membrane and the PM of many cells (Fig. 4.4). This change in the intracellular localization pattern suggests that

Rho1 play a role in the oxidative stress response, possibly on the vacuolar membrane, but the physiological significance behind this localization remains to be determined.

118 Rho1 may function through ABC transporter in response to oxidative stress.

Recent work from the Susan Michaelis lab indicates that Rho1 functions in concert with Ycf1 (an ABC family transporter that resides on the vacuolar membrane) to confer resistance to heavy metal toxicity (Paumi et al., 2007). Ycf1 is a vacuolar glutathione S- conjugate transporter that has a role in detoxifying metals in yeast. This study showed that

Tus1, a GEF for Rho1 activates Ycf1, most likely in a Rho1-dependent manner and that this function was necessary for conferring resistance to cadmium and arsenite (Paumi et al.,

2007). Since heavy metal is also known to induce oxidative stress (Brennan and Schiestl,

1996; Liu et al., 2005), I examined the growth phenotype of tus1Δ and ycf1Δ strains in the

presence of H2O2. While tus1Δ conferred H2O2 sensitivity ycf1Δ strains exhibited very mild

sensitivity to H2O2 (Fig. 4.5). I also examined the growth phenotype of other ABC transporter (that reside on the vacuolar membrane) such as bpt1Δ and ybt1Δ mutants in the

presence of H2O2. Deletion of these ABC transporters conferred mild sensitivity to H2O2

((Fig. 4.5).

DISCUSSION

Rho1 is an essential member of the Rho family of GTPases that has been shown to regulate a number of cellular processes. Here I tested the role of Rho1 in the cell's response to oxidative stress. Several of the rho1ts mutants exhibited different degrees of sensitivity to various oxidants, suggesting that specific mutations in the Rho1 protein can interfere with its function required in protecting the cells from oxidative stress. The phenotype is unlikely due to the general lack of adequate Rho1 functioning because different rho1ts mutants exhibited varying degree of sensitivity to different oxidants. 119 Since recent studies suggested the involvement of Pkc1-MAPK in different stress responses (including oxidative stress), I tested the growth phenotype of some deletion mutants for genes involved in the Pkc1-MAPK pathway (Vilella et al., 2005). I found that mpk1Δ mutant was mildly sensitive to H2O2. This data suggests that different components of this pathway may be responding to external oxidants differently, and this oxidant-mediated cell signaling may not be a linear top-down pathway. Surprisingly, none of the pkc1 mutants were sensitive to H2O2, which contradicts a recent report from another lab (Vilella et al., 2005). The pkc1 mutant phenotype to oxidative stress could be specific to their lab strain background.

GFP-Rho1 enriched on the vacuolar membrane of many cells after H2O2 treatment, suggesting that Rho1 could be functioning on the vacuolar membrane in cells undergoing oxidative stress. Since specific rho1ts mutants showed varying degrees of sensitive growth phenotype to different oxidants, it will be interesting to test their localization prior to and after H2O2 treatment.

Interestingly, I found that ycf1Δ conferred H2O2 sensitivity, while the deletion of other ABC transporter genes exhibited a mild sensitive growth phenotype in the presence of H2O2 (data not shown). There is a good possibility that Rho1 could be functioning through more than one ABC transporter to facilitate cell survival following oxidative stress. This could explain the mild sensitivity of individual ABC transporter mutants.

Future studies testing the growth phenotype of cells completely lacking the ABC transporter function on the vacuolar membrane will help us understand their function better. Additional experiments testing if overexpression of ABC transporters can rescue

120 the sensitive growth phenotype of some rho1ts mutants to oxidants, and, testing if Rho1 interacts with any of these transport proteins will be useful.

121 122 123 124 125 126 APPENDIX: B

TESTING THE POSSIBLE BENEFITS OF Rho5-MEDIATED CELL DEATH IN

AGING CULTURES.

As discussed in Chapter 2, the presence of Rho5 confers sensitivity to different oxidants (such as H2O2, DEM and paraquat). Rho5 is involved in the H2O2-mediated cell death process. These observations raise a question about the possible benefits of the

Rho5-mediated cell death program in yeast. I speculated that the Rho5-mediated cell death program might have evolved to facilitate the removal of old, deteriorating cells from a population of cell that has to compete and grow consuming the limited nutrients available in the environment. This process of getting rid of the old, unhealthy cells would then allow the growth of younger and healthier cells, which can consume and maximize the benefits of the dwindling nutrients. In addition to the above explanation, other possibilities such as Rho5 may be participating in some cellular function that is not known at present, or that the rho5Δ mutants may be unable to repair the cellular damages

(e.g. DNA damage), resulting in low cell survival ability cannot be ruled out.

Cell Survival Competition Assay

Yeast strains used in this study are listed in the Strain Table (Table 1).

127 To compare cell survival between WT and rho5Δ cells, a competition assay was carried out by co-culture. Equal number of WT (HYP210), and rho5Δ cells marked with kanamycin resistant gene deletion cassette (HPY720) were co-cultured in YPD. Aliquots of this culture were taken at specific intervals (every 5th day for the first 30 days followed by every 2nd day till 42 days of co-culturing) and plated on YPD plates and

YPD-Geneticin plates to distinguish between the two strains, and to count the number of colonies formed from each strain.

As a control a similar cell survival growth competition assay was carried out for

WT strain (HPY211) and gic2Δ strain (HPY1470), which is marked with kanamycin resistant gene deletion cassette.

The percentage of WT and rho5Δ cells surviving in the culture was determined for a period of 42 days (Fig. 5.1). After the initial period of similar growth, the WT cells prevailed over the rho5Δ cells (Fig. 5.1), particularly after 30 days of co-culturing. These trends were consistently observed in three independent sets of experiment. These data suggest that rho5Δ cells, which lack the Rho5-mediated cell death machinery, cannot efficiently get rid of the old damaged cells. As a result they may not be able to survive when they have to compete with the WT cells, which have the functional Rho5-mediated cell death program. Such a difference of cell survival was not observed in the strain used as a control (Fig. 5.2).

DISCUSSION

As described in chapter 2, Rho5 conferred sensitivity to different oxidants such as

H2O2, paraquat and DEM. Rho5 is also necessary for apoptotic cell death. Here I 128 addressed the question of why the yeast cells have retained the Rho5-mediated cell death program?

Many studies in other labs have suggested that ROS mediated accumulation of oxidative damage is a critical aspect of aging (for review see Kregel and Zhang, 2006).

Based on the current understanding of the impact of ROS on cellular function, I speculated that Rho5 may be acting as a sensor that will selectively trigger cell death in old/damaged cells. This process is beneficial to the overall population of cells competing for limited nutrients, as the old deteriorating population of cells will be eliminated leaving behind younger cells that can effectively utilize the nutrients and proliferate.

Consistent with the idea, the presence of Rho5 is likely to provide long-term advantage to the cells in aging cultures, since a control strain gic2Δ does not show such phenotype.

The WT cells grew better than rho5Δ cells. Similar cell survival competition experiments using rho5Δ strain carrying a different marker and in a different strain background will be a useful control.

129 130 APPENDIX C

A CANDIDATE APPROACH TO IDENTIFY DOWNSTREAM TARGET OF

Rho5.

As discussed in chapter 2, Rho5 GTPase is necessary for H2O2-induced cell death in budding yeast. Since Rho5-mediated cell death accompanied excessive ROS accumulation and DNA fragmentation (similarly to mammalian cells undergoing apoptosis), I speculated that Rho5 could be using genes known to function in apoptosis as downstream targets to mediate cell death. Studies in other labs have identified proteins such as Yca1 (metacaspase) (Madeo et al., 2002, Khan et al., 2005), Aif1 (apoptosis- inducing factor) (Wissing et al., 2004), Ste 20 (Ahn et al., 2005), and Nuc1

(endonuclease G) (Buttner et al., 2007), as mediators of apoptotic cell death in yeast.

Hence I tested if any of the above-mentioned genes is acting as a downstream effector of

Rho5.

The laboratory WT strains exhibited varying degrees of sensitivity to H2O2 depending on the background. HPY210 (WT used in this study) was less sensitive to 1- 2

mM H2O2 than BY4742 (from Open Biosystems) - the WT strain in the background used by the Saccharomyces Genome Deletion Consortium. This difference may explain why trr1 deletion causes lethality in the BY4742 background, unlike in the HPY210 strain

131 background. Despite the differences, the rho5Δ mutant in the BY4742 background was

more resistant to 1 mM H2O2 than BY4742 (Fig. 6.1 & 6.2; Fig. 2.4).

Since the laboratory WT strains exhibited varying degrees of sensitivity to H2O2 depending on the background, all the deletion strains used in this study were constructed in HPY210 background and were transformed with plasmid expressing Rho5G12V mutant

protein. Deletion mutants that do not exhibit cell death after H2O2 treatment are likely to act in concert with Rho5 in mediating H2O2-induced cell death process.

Construction of yeast deletion strains

Primers used in this study are listed in the Primer Table (Table 3).

Yeast deletion library was purchased from the Open Biosystem consisting of

~4500 yeast deletion strains, each carrying a single gene deletion. Each gene deleted is replaced with a cassette carrying a kanamycin resistance gene, which would make the deletion strain resistant to kanamycin.

To construct a yca1Δ strain in HPY210 background, yca1Δ::kanMX4 cassette was

PCR amplified using primers oYCA11 and oYCA12 and genomic DNA extracted from yca1Δ strain (BY4742 strain background) as template. The resulting PCR fragment was transformed into YEF473 (a/α) diploid cells (Bi and Pringle, 1996) by one-step gene disruption (Rothstein, 1991), yielding yca1Δ/YCA1 heterozygous diploid strain, which was verified by genomic PCR. After sporulation, among the meiotic progeny yca1Δ haploid strain was isolated (based on their ability to grow on G-418 containing YPD plates).

132 Similarly, aif1Δ, yca1Δ, nuc1Δ, and, ste20Δ strains were constructed in HPY210 background as described above, except that primer pairs of oAIF11 & oAIF12, oSTE201

& oSTE202, oNUC11 & oNUC12, and oFRE11 & oFRE12 were used.

Yeast strains and their phenotype in the presence of oxidants.

All the haploid deletion strains constructed in HPY210 background were transformed with rho5G12V plasmid (pHP1282) and empty vector (pRS426; pHP685). To

G12V test the growth phenotype in the presence of H2O2, cells carrying the rho5 plasmid and the empty vector were grown to mid logarithmic phase culture (in SD-URA media) were then diluted to OD600 = 1, and then serially diluted (5-fold). 200 µl of each culture was transferred to a 96-well plate in the presence or absence of 4.4mM H2O2 and no oxidant. The cells were incubated for 4hrs at 30°C, then plated on SD-URA plates using sterile metal pins. A rho5Δ cells carrying each RHO5 plasmid (RHO5WT, rho5G12V and rho5K16N) and the empty vector were also tested as described above, as control.

G12V The aif1Δ strain carrying the rho5 plasmid was hypersensitive to H2O2 similar to the rho5Δ cells carrying the rho5G12V plasmid (Fig. 6.3), suggesting that Aif1 may not be the downstream target of Rho5. Similarly, yca1Δ, nuc1Δ, and ste20Δ strain, each

G12V carrying the rho5 plasmid was hypersensitive to H2O2 similar to the rho5Δ cells carrying the rho5G12V plasmid (Fig. 6.4, 6.5 and 6.6, respectively).

Contrary to our expectation, Rho5 is unlikely to use proteins that are known to participate in the process of apoptosis to mediate H2O2-induced cell death. This study suggests that the Rho5 mediated cell death process uses targets, which are distinct from the yeast apoptotic machinery reported in other labs. Since Rho5 mediated cell death 133 accompanies ROS accumulation and vacuolar defect (see Chapter 2), there is a good possibility that Rho5 could be interfering with proteins involved in either vacuolar function, vesicular trafficking, redox homeostasis maintenance or mitochondrial function.

Future studies identifying downstream targets of Rho5 will shed more light on the mechanism of this cell death process.

Alternatively, the possibility that Yca1, Aif1, Nuc1 and Ste20 could be acting upstream of Rho5 in mediating cell death cannot be ruled out. If the above is true, then the deletion of the above-mentioned gene should give a resistant growth phenotype to

H2O2. Since yca1Δ, aif1Δ, nuc1Δ, and ste20Δ strains did not exhibit resistance to H2O2

(data not shown), it is unlikely that these genes are acting upstream of Rho5 in the H2O2- induced cell death pathway.

134 135 136 137 138 139 Strain Relevant Genotypea Source

YEF473 a/α trp1-Δ63 leu2-Δ1 ura3-52 his3-Δ200 lys2-801amber Bi and Pringle (1996)

HPY210 a Segregant from YEF473

HPY211 α Segregant from YEF473

b HPY713 a/α RHO5/rho5Δ::kanMX4 See text (PJK)

HPY720 a rho5Δ::kanMX4 This study (PJK) HPY999 a rho5Δ::kanMX4 rho5G12V::URA3 This study

HPY1156 a rho5Δ::kanMX4 RHO5-GFP::URA3 This study

HPY1157 a rho5Δ::kanMX4 rho5G12V-GFP::URA3 This study

HPY1159 a rho5Δ::kanMX4 TRR1-Cherry::TRP1 This study

HPY1160 a TRR1-Cherry::TRP1 This study

HPY1161 a rho5Δ::kanMX4 rho5G12V::URA3 TRR1-Cherry::TRP1 This study

HPY1162 a rho5Δ::kanMX4 TRR1-YFP(N)::TRP1 This study

HPY1163 a TRR1-YFP(N)::TRP1 RHO5-YFP(C)::URA3::rho5Δ::kanMX4 This study

HPY1164 a TRR1-YFP(N)::TRP1 rho5G12V-YFP(C)::URA3::rho5Δ::kanMX4 This study

HPY1165 a TRR1-YFP(N)::TRP1 rho5K16N-YFP(C)::URA3::rho5Δ::kanMX4 This study

HPY1175 a trr1Δ::HIS3 This study

HPY1176 a/α TRR1/trr1Δ::HIS3 This study

HPY1185 a rho5Δ::kanMX4 rho5C328S-GFP::URA3 This study

HPY1186 a rho5Δ::kanMX4 rho5K321-325S-GFP::URA3 This study

HPY1470 a gic2Δ::kanMX4 Kawasaki et al. (2003)

HPY1506 a rho5Δ::kanMX4 rho5-RITC-GFP::URA3 This study

Table 1. Yeast strains used in this study

Continued 140 Table 1. continues

a EGY48* α his3 trp1 ura3-52 leu2::pLEU2-LexAop6 Gyuris et al. (1993)

a BY4742* α his3Δ1 lys2Δ0 ura3Δ0 Open Biosystems

a HPY657* α rho5Δ::kanMX4 Open Biosystems

a HPY16* a his3-Δ1 leu2 trp1- Δ63 ura3-52 prb1-1122 pep4-3 prc1-40 Park et al. (1993)

HPY910* a TRR1-13MYC-kanMX6 This study (EA)ac

a* All strains except those marked with asterisk (*) are congenic to YEF473 except as indicated. HPY910 is congenic to HPY16 except as indicated. HPY657 was purchased from the Open biosystems and is congenic to BY4742 except as indicated. bPJK: Strains constructed by Dr. Pil Jung Kang cEA: Strains constructed by Elizabeth Angerman

References

Bi, E., and Pringle, J.R. (1996). ZDS1 and ZDS2, genes whose products may regulate Cdc42p in Saccharomyces cerevisiae. Mol. Cell. Biol. 16, 5264–5275.

Gyuris, J., Golemis, E., Chertkov, H., Brent, R. (1993). Cdi1, a human G1 and S phase protein phosphatase that associates with Cdk2. Cell. 75, 791-803.

Park, H.-O., Chant, J., and Herskowitz, I. (1993). BUD2 encodes a GTPase activating protein for Bud1/Rsr1 necessary for proper bud-site selection in yeast. Nature. 365, 269–274.

141 Plasmid Description Source pRS426 URA3 (high copy) Christianson et al (1992) pRS314 TRP1 (CEN) Sikorski and Hieter (1989) pRS316 URA3 (CEN) Sikorski and Hieter (1989) pRS306 URA3 (integrative) Sikorski and Hieter (1989) pRS304 TRP1 (integrative) Sikorski and Hieter (1989) pFA6a-13Myc-kanMX6 Longtine et al. (1998) pSG14-URA3 RHO5 in pRS316 H. Madhani pSG14V12-306 rho5G12V in pRS306 H. Madhani pGEX2T-HA-RHO5 Roumanie et al. (2001) pYN-1 YFP–N (1-158AA) NotI-XbaI fragment in Kan+ plasmid D. Bisaro pYC-1 YFP-C (158-238AA) NotI-XbaI fragment in Kan+ plasmid D. Bisaro pHP767 GFP-BUD1 in YCp50 Park et al. (2002) pHP835 BUD5-HA3 in YCp50 Kang et al. (2004) pHP1280 RHO5 in pRS426. This study pHP1281 rho5G12V in pRS426. This study pHP1282 rho5K16N in pRS426 This study (PJK)d pHP1382 RHO5 in pEG202 This study pHP1383 TRR1 in pJG4-5 This study pHP1386 rho5G12V in pEG202 This study pHP1387 rho5K16N in pEG202 This study pHP1463 mCherry in pBluescript This study (PJK)d

Table 2. Plasmids used in this study

Continued 142 Table 2. continues pHP1465 TRR1 in pRS426 This study pHP1466 trr1Δ::HIS3 in pRS426 This study pHP1467 RHO5-GFP in pRS306 This study pHP1476 pRS426 lacking NotI This studya pHP1477 pRS426 lacking BamHI This studyb pHP1492 NotI was introduced in TRR1 before Stop codon in pRS426 This study pHP1494 TRR1-mCherry in pRS304 This study pHP1496 TRR1-HA3 in pRS314 This study pHP1499 RHO5-GFP in pRS426 This study pHP1500 RHO5-YFP-C in pRS426 This study pHP1501 RHO5-YFP-C in pRS306 This study pHP1502 rho5G12V (lacking a.a 248-316) in pRS426 This study pHP1503 rho5G12V-GFP in pRS426 This study pHP1504 rho5G12V-YFP-C in pRS426 This study pHP1505 rho5G12V-GFP in pRS306 This study pHP1506 rho5G12V-YFP-C in pRS306 This study pHP1507 rho5K16N (lacking a.a 248-316) in pRS426 This study pHP1509 rho5K16N-YFP-C in pRS426 This study pHP1511 rho5K16N-YFP-C in pRS306 This study pHP1512 TRR1-mCherry in pRS426 This study pHP1513 TRR1-YFP-N in pRS426 This study pHP1514 TRR1-YFP-N in pRS304 This study pHP1517 pRS304 derivative digested with XhoI, SalI and self-ligated This studyc

Continued

143 Table 2. Continues pHP1530 pRP GFP-ras/RitC Onken et al. (2006) pHP1535 rho5C328S (lacking a.a 248-316) in pRS426 This study pHP1536 rho5K321-325S (lacking a.a 248-316) in pRS426 This study pHP1537 rho5C328S -GFP in pRS426 This study pHP1538 rho5K321-325S-GFP in pRS426 This study pHP1539 rho5C328S -GFP in pRS306 This study pHP1540 rho5K321-325S-GFP in pRS306 This study pHP1544 RHO5 (lacking a.a. 248-316) in pRS426 This study pHP1570 TRR1-mCherry in pRS314 This study pHP1669 RitC (148-209 a.a domain) &PCR product of

oligos oRHO525, oRHO52 in pRS306 This Study pHP1670 rho5 (lacking a.a. 315-332) fused to RitC in pRS426 This study pHP1671 rho5Δ (18a.a.)-RitC-GFP in pRS426 This study pHP1672 rho5Δ (18a.a.)-RitC-GFP in pRS306 This study

a NotI site in pRS426 was knocked out by Klenow filling b BamHI site in pRS426 was knocked out by Klenow filling c pRS304 was digested with XhoI, SalI and self-ligated to knock out the restriction sites. dPJK: Plasmids made by Dr. Pil Jung Kang

144 Primer name Sequence oRHO51 ACTCGGATCCTGCCATGTCCTGCTGCTTAC oRHO52 ACTCCTGATCAAATAGCCTACGCCGCG oRHO55 GGTGATGGTGCAGTAGGTAATACGTCACTTCTAATATCATATAC oRHO56 GTATATGATATTAGAAGTGACGTATTACCTACTGCACCATCACC oRHO59 GCGGGGATCCTGATGAGGTCTATTAAATGTGTG oRHO510 GCGGCTCGAGACTTTGACTTCTTTTTCTTCTTGTC oRHO512 CCTGCAGCCCGGGCCTTTCTCTCCCCTGAATC oRHO515 GAAGGGCGGCCGCACGAGAAACGACAAGAAG oRHO516 GAAGCGCGGCCGCAATCTCCATTGGTGTTGGT oRHO517 GCGGGAGCTCTTCTTCAAAGTCAAAGTGTGTAATAC oRHO518 GAGGGAGCTCGAAGAGTCGTTTCTCGTGCGGCCGC oRHO519 GGACTCTAGAATCAAGGAGGCAGAAAAAG oRHO520 GGACTCTAGAGTATTACAGACTTTGACTTCTTTTTCTTC oRHO523a GAACGCGGCCGCAAAAATATTCGAGAACAAAAACAG oRHO524a GAACTCTAGAAGCTTGCTTGTTGTGATGATTATTTG oRHO525 ACTCGAGCTCTAATATCAAGGAGGCAGAAAAAG oTRR11 TCTGCTGGCTCTGGTTGTATGGCCGCTTTGGATGCTGAGAAATACTTAACTTCCC TAGAACGGATCCCCGGGTTAATTAA oTRR12 CGATTTGTTGGATAAGTATACAAAAATTTGAGTGTATCTATTTTATAATGGAAAATTC ATGAATTCGAGCTCGTTTAAAC

Table 3. Oligonucleotides/Primers used in this study

Continued 145 Table 3. continues oTRR15 CCGCGAATTCATGGTTCACAACAAAGTTACTATC oTRR16 CCGCCTCGAGCTATTCTAGGGAAGTTAAGTATTTCTC oTRR17 CAGTGCAGCGCAACTAGTGACAGAACGTAC oTRR18 GCGGACTAGTAGGTCTGTGCTACTCAATTGC oTRR19 GCGGGGATCCTATTGATAATATGATGTTTACTTTCG oTRR110 GCGGGGATCCACTTCCCTAGAATAGATGAATTTTCC oTRR113 GGGAGCGGCCGCTAGATGAATTTTCCATTATAAAATAGATAC oTRR114 GGGAGCGGCCGCATTCTAGGGAAGTTAAGTATTTCTC oTRR115 GTAAAGGATTTCAATTTTTTCGTTCTTCTCAGCACGCTTTTGCATAATGGT o18s1748 TAATGATCCTTCCGCAGGTTCACCTACGGAAACCTTGTTACGAC oYFP1 GGGAGCGGCCGCATGGTGAGCAAGGGCGAGGAG oYFP2 GGGAGCGGCCGCTAGCCTTCTGCTTGTCGGCCATGAT oYFP3 GGGAGCGGCCGCAACGGCATCAAGGTGAACTTC oYFP4 GGGAGCGGCCGCTGATAGATCTCTTGTACAGCTC

ORED1CHERRY GCAGGGCGGCCGCGTGAGCAAGGGCGAGGAGGATAAC

ORED2CHERRY GCACGGCGGCCGCCCTTGTACAGCTCGTCCATGCCGCC oAIF11 GCGGCTCGAGCTGCGCTATCTCGTGCCCTCG oAIF12 GCGGCCCGGGGGCATCACGCCTATACCTTAACTG oYCA11 GCACCCCGGGCTGGTACCCTAACGCTAGCG oYCA12 GCACCTCGAGCGAATTAGACGTAACTGTCCC oNUC11 GCACCCCGGGCAGGAACTGAATCAGTGCGC oNUC12 GCACCTCGAGGGGCGTGCGCAGAACTATACC oSTE201 GCACCTCGAGGGATTGTGGGATCTCACCGCG

Continued 146 Table 3. continues oSTE202 CACAATACCCGGGTTTCGTGCGACTATGTC

147 REFERENCES

Abeliovich, H. & Klionsky, D. J. 2001. Autophagy in yeast: mechanistic insights and physiological function. Microbiol Mol. Biol. Rev. 65, 463-479.

Abo, A., Pick, E., Hall, A., Totty, N., Teahan, C.G., Segal, A.W. 1991. Activation of the NADPH oxidase involves the small GTP-binding protein p21rac1. Nature. 353(6345): 668-70. Links

Ahn, S-H., Cheung, W. L., Hsu, J. Y., Diaz, R. L., Smith, M. M., Allis, C. D. 2005. Sterile 20 Kinase Phosphorylates Histone H2B at Serine 10 during Hydrogen Peroxide- Induced Apoptosis in S. cerevisiae. Cell. 120: 25-36.

Alic, N., Higgins, V. J., Dawes, I. W. 2001. Identification of a Saccharomyces cerevisiae gene that is required for G1 arrest in response to the lipid oxidation product linoleic acid hydroperoxide. Mol. Biol. Cell. 12: 1801-1810.

Annan, R.B., Wu, C., Waller, D.D. Whiteway, M., Thomas, D.Y. 2008. Rho5p is involved in mediating the osmotic stress response in S. cerevisiae and its activity is regulated via Msi1p and Npr1p by phosphorylation and ubiquitination. Eukaryot. cell.

Ausubel, F.M., R. Brent, R.E. Kingston, D.D. Moore, J.G. Seidman, and K. Struhl. 1999. Current protocols in moleuclar biology. John Wiley & Sons, Inc.

Bi, E., and J.R. Pringle. 1996. ZDS1 and ZDS2, genes whose products may regulate Cdc42p in Saccharomyces cerevisiae. Mol. Cell. Biol. 16:5264-5275.

Bivona, T. G., Quatela, S. E., Bodemann, B. O., Ahearn, I. M., Soskis, M. J., Mor, A., Miura, J., Wiener, H. H., Wright, L., Saba, S. G., Yim, D., Fein, A., Pérez de Castro, I., Li, C., Thompson, C. B., Cox, A. D. & Philips, M. R. 2006. PKC regulates a farnesyl- electrostatic switch on K-Ras that promotes its association with Bcl-XL on mitochondria and induces apoptosis. Mol. Cell. 21, 481-493.

Branco, M.R., Marinho, H.S., Cyrne, L., Antunes, F. 2004. Decrease of H2O2 plasma membrane permeability during adaptation to H2O2 in Saccharomyces cerevisiae. J. Biol. Chem. 279 (8): 6501-6506.

Brandes, R.P., Kreuzer, J. 2005. Vascular NADPH oxidases: molecular mechanisms of activation. Cardiovasc. Res. 65 (1): 16-27.

148 Brennan RJ, Schiestl RH. 1996. Cadmium is an inducer of oxidative stress in yeast. Mutat. Res. 356 (2): 171-178.

Buttner, S., Eisenberg, T., Herker, E., Carmona-Gutierrez, D., Kroemer, G., Madeo, F. 2006. Why yeast cells can undergo apoptosis: death in times of peace, love, and war. J. Cell. Biol. 175: 521-525.

Büttner, S., Eisenberg, T., Carmona-Gutierrez, D., Ruli, D., Knauer, H., Ruckenstuhl, C., Sigrist, C., Wissing, S., Kollroser, M., Fröhlich, K.-U., Sigrist, S. & Madeo, F. 2007. Endonuclease G regulates budding yeast life and death. Mol. Cell. 25, 233-246.

Cai, H. 2005. Hydrogen peroxide regulation of endothelial function: Origins, mechanisms and consequences. Cardiovasc. Res. 68: 26-36.

Carmel-Harel, O., Stearman, R., Gasch, A. P., Botstein, D., Brown, P. O., Storz, G. 2001. Role of thioredoxin reductase in the Yap1p-dependent response to oxidative stress in Saccharomyces cerevisiae. Mol. Microbiol. 39: 595-605.

Casey, P.J., Seabra, M.C. 1996. Protein prenyltransferases. J. Biol. Chem. 271(10): 5289- 5292.

Causton, H.C., Ren, B., Koh, S.S., Harbison, C.T., Kanin, E., Jennings, E.G., Lee, T.I., True, H.L., Lander, E.S., Young, R.A. 2001. Remodeling of Yeast Genome Expression in Response to Environmental Changes. Mol Biol Cell. 12 (2): 323-337.

Chae, H.Z., Chung, S.J., Rhee, S.G. 1994. Thioredoxin-dependent peroxide reductase from yeast. J. Biol. Chem. 269: 27670-27678.

Chan, T.-F., Carvalho, J., Riles, L. & Zheng, X. F. S. 2000. A chemical genomics approach toward understanding the global functions of the target of rapamycin protein (TOR). Proc. Natl Acad. Sci. USA97, 13227–13232.

Christianson, T.W., R.S. Sikorski, M. Dante, J.H. Shero, and P. Hieter. 1992. Multifunctional yeast high-copy-number shuttle vectors. Gene. 100:119-122.

Cohen, G.M.. 1997. Caspases: the executioners of apoptosis. Biochem J. 326: 1-16.

Cole, K. C., McLaughlin, H.W., Johnson, D.I. 2007. Use of Bimolecular Fluorescence Complementation To Study In Vivo Interactions between Cdc42p and Rdi1p of Saccharomyces cerevisiae. Ekuaryotic. Cell. 6: 378-387.

De Minicis, S., Brenner, D.A. 2007. NOX in liver fibrosis. Arch. Biochem. Biophys. 462 (2): 266-272.

149 Dever, T.E., Glynias, M.J., Merrick, W.C. 1987. GTP-binding domain: three consensus sequence elements with distinct spacing. Proc. Natl. Acad. Sci. 4 (7): 1814-1818.

Devine, S. E. & Boeke, J. D. 1994. Efficient integration of artificial transposons into plasmid targets in vitro: a useful tool for DNA mapping, sequencing and genetic analysis. Nucleic Acids Res. 22, 3765–3772.

Diekmann, D., Abo, A., Johnston, C., Segal, A. W. & Hall, A. 1994. Interaction of Rac with p67phox and regulation of phagocytic NADPH oxidase activity. Science. 265: 531- 533.

Dix, M.M., Simon, G.M., Cravatt, B.F. 2008. Global mapping of the topography and magnitude of proteolytic events in apoptosis. Cell. 134 (4): 679-691.

Ellis, R.E., Yuan, J.Y., Horvitz, H.R. 1991. Mechanisms and functions of cell death. Annu Rev Cell Biol.7: 663-698.

Enari, M., Sakahira, H., Yokoyama, H., Okawa, K., Iwamatsu, A., Nagata, S. 1998. A caspase-activated DNase that degrades DNA during apoptosis, and its inhibitor ICAD. Nature. 391 391 (6662): 43-50

Etienne-Manneville, S., Hall, A. 2002. Rho GTPases in cell biology. Nature. 420: 629- 635.

Freeman, J.L., Abo, A., Lambeth, J.D. 1996. Rac "insert region" is a novel effector region that is implicated in the activation of NADPH oxidase, but not PAK65. J. Biol. Chem. 271 (33): 19794-19801.

Fukumoto, Y., Kaibuchi, K., Hori, Y., Fujioka, H., Araki, S., Ueda, T., Kikuchi, A., Takai, Y. 1990. Molecular cloning and characterization of a novel type of regulatory protein (GDI) for the rho proteins, ras p21-like small GTP-binding proteins. Oncogene. 5: 1321–1328.

Gan, Z.R. 1991. Yeast thioredoxin genes. J. Biol. Chem. 266 (3): 1692-1696.

Garcia-Ranea, J. A., Valencia, A. 1998. Distribution and functional diversification of the ras superfamily in Saccharomyces cerevisiae. FEBS. letters. 434: 219-225.

Gasch, A.P., Spellman, P.T., Kao, C.M., Carmel-Harel, O., Eisen ,M.B., Storz, G., Botstein, D., Brown, P.O. 2000. Genomic expression programs in the response of yeast cells to environmental changes. Mol. Biol. Cell. 11: 4241-4257.

Gasdaska, J.R., Harney, J.W., Gasdaska, P.Y., Powis, G., Berry, M.J. 1999. Regulation of human thioredoxin reductase expression and activity by 3'-untranslated region selenocysteine insertion sequence and mRNA instability elements. J. Biol.Chem. 274:

150 25379-25385.

Geyer, M., Wittinghofer, A. 1997. GEFs, GAPs, GDIs and effectors: taking a closer (3D) look at the regulation of Ras-related GTP-binding proteins. Curr. Opin. Struct. Biol. 7 (6): 786-792.

Glomset, J.A., Farnsworth, C.C. 1994. Role of protein modification reactions in programming interactions between ras-related GTPases and cell membranes. Annu. Rev. Cell. Biol. 10: 181-205.

Goffeau, A. et al. Life with 6000 genes. 1996. Science274, 546, 563–567.

Grant, C. M. 2001. Role of the glutathione/glutaredoxin and thioredoxin systems in yeast growth and response to stress conditions. Mol. Microbiol. 3: 533-541.

Guthrie, C., and G.R. Fink. 1991. Guide to yeast genetics and molecular biology.

Gyuris, J., E. Golemis, H. Chertkov, and R. Brent. 1993. Cdi1, a human G1 and S phase protein phosphatase that associates with Cdk2. Cell. 75:791-803.

Hall, A. 1990. The cellular functions of small GTP-binding proteins. Science. 249: 635–640.

Hancock, J.F., Magee, A.I., Childs, J.E., Marshall, C.J. 1989. All ras proteins are polyisoprenylated but only some are palmitoylated. Cell. 7: 1167-1177.

Hardwick, J. M., Cheng, W. C. 2004. Mitochondrial programmed cell death pathways in yeast. Dev. Cell. 7: 630-632.

Harrison, J.C., Zyla, T.R., Bardes, E.S., Lew, D.J. 2004. tress-specific activation mechanisms for the "cell integrity" MAPK pathway. J. Biol. Chem. 279(4): 2616-2622

Herker, E., Jungwirth, H., Lehmann, K. A., Maldener, C., Frohlich, K-U., Wissing, S., Buttner, S., Fehr, M., Sigrist, S., Madeo, F. 2004. Chronological aging leads to apoptosis in yeast. J. Cell. Biol. 164: 501-507.

Higuchi Y. 2003. Chromosomal DNA fragmentation in apoptosis and necrosis induced by oxidative stress. Biochem Pharmacol. 66 (8): 1527-1535.

Higuchi, Y., Matsukawa, S. 1997. Appearance of 1~2 Mbp Giant DNA Fragments as an Early Common Response Leading to Cell Death Induced by Various Substances That Cause Oxidative Stress. Free Radical Biology and Medicine, Volume 23(1): 90-99.

Ho, Y., Gruhler, A., Heilbut, A., Bader, G.D., Moore, L., Adams, S-L., Millar, A.,

151 Taylor, P., Bennett, K., Boutilier, K., Yang, L., Wolting, C., Donaldson, I., Schandorff, S., Shewnarane, J., Vo, M., Taggart, J., Goudreault, M., Muskat, B., Alfarano, C., Dewar, D., Lin, Z., Michalickova, K., Willems, A.R., Sassi, H., Nielsen, P.A., Rasmussen, K.J., Andersen, J.R., Johansen, L.E., Hansen, L.H., Jespersen, H., Podtelejnikov, A., Nielsen, E., Crawford, J., Poulsen, V., Sorensen, B.D., Matthiesen, J., Hendrickson, R.C., Gleeson, F,, Pawson, T., Moran, M.F., Durocher, D., Mann, M., Hogue, C.W., Figeys, D., Tyers, M. 2002. Systematic identification of protein complexes in Saccharomyces cerevisiae by mass spectrometry. Nature 415: 180-183.

Hordijk, P. L. 2006. Regulation of NADPH oxidases: the role of Rac proteins. Circ. Res. 98: 453-462.

Holgrem, A. 1989. Thioredoxin and glutaredoxin systems. J.Biol.Chem. 264, 13963–13966.

Hu, C.D., Chinenov, Y., Kerppola, T.K. 2002. Visualization of interactions among bZIP and Rel family proteins in living cells using bimolecular fluorescence complementation. Mol. Cell. 9: 789-798.

Hu, C.D., Kerppola, T.K. 2005. Direct visualization of protein interactions in living cells using bimolecular fluorescence complementation analysis. Protein-Protein Interactions (ed. P. Adams and E. Golemis), Cold Spring Harbor Laboratory Press.

Jamieson, D.J. 1998. Oxidative stress responses of the yeast Saccharomyces cerevisiae. Yeast. 14 (16): 1511-1527.

Jin, C., Reed, J. C. 2002. Yeast and apoptosis. Nature Reviews. 3: 453-459.

Jones, E. W., Webb, G. C. & Hiller, M. A. (1997) in The Molecular and Cellular Biology of the Yeast Saccharomyces, eds. Pringle, J. R., Broach, J. & Jones, E. W. (Cold Spring Harbor Laboratory Press, Plainview, New York), Vol. 3, pp. 363-470.

Kamada, Y., Jung, U.S., Piotrowski, J., Levin, D.E. 1995. The protein kinase C-activated MAP kinase pathway of Saccharomyces cerevisiae mediates a novel aspect of the heat shock response. Genes. Dev. 9: 1559-1571.

Kamata, H., Hirata, H. 1999. Redox regulation of cellular signalling. Cell. Signal. 11: 1- 14.

Kang, P.J., E. Angerman, K. Nakashima, J.R. Pringle, and H.-O. Park. 2004. Interactions among Rax1p, Rax2p, Bud8p, and Bud9p in marking cortical sites for bipolar bud-site selection in yeast. Mol. Biol. Cell. 15:5145-5157.

Kang, P. J., Sanson, A., Lee, B., Park, H-O. 2001. A GDP/GTP exchange factor involved in linking a spatial landmark to cell polarity. Science. 292: 1376-1378.

152 Kawasaki, R., Fujimura-Kamada, K., Toi, H., Kato, H., Tanaka, K. 2003. The upstream regulator, Rsr1p, and downstream effectors, Gic1p and Gic2p, of the Cdc42p small GTPase coordinately regulate initiation of budding in Saccharomyces cerevisiae. Genes Cells. 8, 235-250.

Kawasaki, T., Henmi, K., Ono, E., Hatakeyama, S., Iwano, M., Satoh, H., Shimamoto, K. 1999. The small GTP-binding protein rac is a regulator of cell death in plants. Proc. Natl. Acad. Sci. 96: 10922-10926.

Kern, J.C., Kehrer, J.P. 2005. Free radicals and apoptosis: relationships with glutathione, thioredoxin, and the BCL family of proteins. Front Biosci. 10: 1727-1738.

Keyer, K., Imlay, J.A. 1996. Superoxide accelerates DNA damage by elevating free-iron levels. Proc. Natl. Acad. Sci. 93 (24): 13635-13640.

Kregel, K. C., Zhang, H.J. 2007. An integrated view of oxidative stress in aging: basic mechanisms, functional effects, and pathological considerations. Am J Physiol Regul Integr Comp Physiol. 292(1): R18-36.

Kozminski, K.G., Beven, L., Angerman, E., Tong, A.H., Boone, C., Park, H.-O. 2003. Interaction between a Ras and a Rho GTPase couples selection of a growth site to the development of cell polarity in yeast. Mol. Biol. Cell. 14: 4958-4970.

Kumar, A., Snyder, M. 2001. Emerging technologies in yeast genomics. Nat Rev Genet. 2 (4): 302-312.

Lacal, J. C. 1997. Regulation of proliferation and apoptosis by Ras and Rho GTPases through specific phospholipid-dependent signaling. FEBS. letters. 410: 73-77. Lee, K.S., Levin, D.E. 1992. Dominant mutations in a gene encoding a putative protein kinase (BCK1) bypass the requirement for a Saccharomyces cerevisiae protein kinase C homolog. Mol. Cell. Biol. 12(1): 172-82.

Levine, A., Belenghi, B., Damari-Weisler, H., Granot, D. 2001. Vesicle-associated membrane protein of Arabidopsis suppresses Bax-induced apoptosis in yeast downstream of oxidative burst. J. Biol. Chem. 276(49):46284-46289.

Levin, D.E., Fields, F.O., Kunisawa, R., Bishop, J.M., Thorner, J. 1990. A candidate protein kinase C gene, PKC1, is required for the S. cerevisiae cell cycle. Cell. 62(2): 213- 224.

Li, L.Y., Wang, X. Endonuclease G is an apoptotic DNase when released from mitochondria. 412: 95–99.

Lin, S. J., Culotta, V. C.1996. Suppression of oxidative damage by Saccharomyces

153 cerevisiae ATX2, which encodes a manganese-trafficking protein that localizes to Golgi- like vesicles. Mol. Cell. Biol. 11:6303-12.

Liu, J., Zhang, Y., Huang, D., Song, G. 2005. Cadmium induced MTs syn-. thesis via oxidative stress in yeast Saccharomyces cerevisiae. Mol. Cell. Biol. 280: 139–145.

Longtine, M.S., A.I. McKenzie, D.J. Demarini, N.G. Shah, A. Wach, A. Brachat, P. Philippsen, and J.R. Pringle. 1998. Additional modules for versatile and economical PCR-based gene deletion and modification in Saccharomyces cerevisiae. Yeast. 14: 953- 961.

Madeo, F., Frohlich, E., Ligr, M., Grey, M., Sigrist, S. J., Wolf, D. H., Frohlich, K-U. 1999. Oxygen stress: a regulator of apoptosis in yeast. J. Cell. Biol. 145: 757-767.

Madeo, F., Herker, E., Maldener, C., Wissing, S., Lachelt, S., Herlan, M., Fehr, M., Lauber, K., Sigrist, S. J., Wesselborg, S. 2002. A caspase-related protease regulates apoptosis in yeast. Mol. Cell. 9: 911-917.

Mahrus, S., Trinidad, J.C., Barkan, D.T., Sali, A., Burlingame, A.L., Wells, J.A. 2008. Global sequencing of proteolytic cleavage sites in apoptosis by specific labeling of protein N termini. Cell. 134(5): 866-876.

Marelli, M., Smith, J.J., Jung, S., Yi, E., Nesvizhskii, AI., Christmas, R.H., Saleem, R.A., Tam, Y.Y., Fagarasanu, A., Goodlett, D.R., Aebersold, R., Rachubinski, R.A., Aitchison, J.D. 2004. Quantitative mass spectrometry reveals a role for the GTPase Rho1p in actin organization on the peroxisome membrane. J. cell. Biol. 167: 1099-1112.

Mazur, P., Baginsky, W. 1996. In Vitro Activity of 1,3-B-D-Glucan Synthase Requires the GTP-binding Protein Rho1. J Biol Chem 271: 14604-14609.

Michaelson, D., Silletti, J., Murphy, G., D'Eustachio, P., Rush, M., Philips, M.R. 2001. Differential localization of Rho GTPases in live cells: regulation by hypervariable regions and RhoGDI binding. J. Cell. Biol. 152 (1): 111-126.

Morgan, B.A., Banks, G.R., Toone, W.M., Raitt, D., Kuge, S., Johnston, L.H. 1997. The Skn7 response regulator controls gene expression in the oxidative stress response of the budding yeast Saccharomyces cerevisiae. EMBO J. 16: 1035-1044.

Mroczek, S., Kufel, J. 2008. Apoptotic signals induce specific degradation of ribosomal RNA in yeast. Nuc. Acid. Res. 36(9): 2874-2888.

Niedenthal, R.K., L. Riles, M. Johnston, and J.H. Hegeman. 1996. Green fluorescent protein as a marker for gene expression and subcellular localization in budding yeast. Yeast. 12: 773–786.

154 Nobes, C. D., Hall, A. 1995. Rho, Rac and Cdc42 regulate the assembly of multimolecular focal complexes associated with actin stress fibers, lamellipodia and filopodia. Cell. 81, 53–62.

Norbury, C.J., Zhivotovsky, B. 2004. DNA damage-induced apoptosis. Oncogene. 23: 2797-2808.

Onken, B., Wiener, H., Philips, M. R. Chang EC. 2006. Compartmentalized signaling of Ras in fission yeast. Proc. Natl. Acad. Sci. 103(24): 9045-9050.

Park, H.-O., J. Chant, and I. Herskowitz. 1993. BUD2 encodes a GTPase-activating protein for Bud1/Rsr1 necessary for proper bud-site selection in yeast. Nature. 365: 269- 274.

Park, H.-O., P.J. Kang, and A.W. Rachfal. 2002. Localization of the Rsr1/Bud1 GTPase involved in selection of a proper growth site in yeast. J. Biol. Chem. 277: 26721-26724.

Paumi, C.M., Menendez, J., Arnoldo, A., Engels, K., Iyer, K.R., Thaminy, S., Georgiev, O., Barral. Y,, Michaelis, S,. Stagljar, I. 2007. Mapping protein-protein interactions for the yeast ABC transporter Ycf1p by integrated split-ubiquitin membrane yeast two- hybrid analysis. Mol. Cell. 26 (1): 15-25.

Pedrajas, J.R., Kosmidou, E., Miranda-Vizuete, A., Gustafsson, J-A., Wright, A.P.H., Spyrou, G. 1999. Identification and functional characterization of a novel mitochondrial thioredoxin system in Saccharomyces cerevisiae. J. Biol. Chem. 274: 6366-6373.

Pedrajas, J.R., Miranda-Vizuete, A., Javanmardy, N., Gustafsson, J.A., Spyrou, G. 2000. Mitochondria of Saccharomyces cerevisiae contain one-conserved cysteine type peroxiredoxin with thioredoxin peroxidase activity. J. Biol. Chem. 275: 16296-16301.

Raff, M. Cell suicide for beginners. Nature. 396 (6707): 119-122.

Ridley, A.J. 2001. Rho family proteins: coordinating cell responses. Trends. Cell. Biol. 11(12): 471-477.

Ridley, A. J., Hall, A. 1992. The small GTP-binding protein Rho regulates the assembly of focal adhesions and stress fibers in response to growth factors. Cell. 70: 389–399.

Rietsch, A., Beckwith, J. 1998. The genetics of disulfide bond metabolism. Annu. Rev. Genet. 32: 163–184.

Rothstein, R. 1991. Targeting, disruption, replacement, and allele rescue: Integrative DNA transformation in yeast. Methods. Enzymol. 194: 281-301.

Roumanie O, Weinachter C, Larrieu I, Crouzet M, Doignon F. 2001. Functional

155 characterization of the Bag7, Lrg1 and Rgd2 RhoGAP proteins from Saccharomyces cerevisiae. FEBS. Lett. 506: 149-156.

Saka, A., Abe, M., Okano, H., Minemura, M., Qadota, H., Utsugi, T., Mino, A., Tanaka, K., Takai, Y., Ohya, Y. 2001. Complementing yeast rho1 mutation groups with distinct functional defects. J. Biol. Chem. 276: 46165-46171.

Shaner, N.C., Campbell, R.E., Steinbach, P.A., Giepmans, B.N., Palmer, A.E., Tsien, R.Y. 2004. Improved monomeric red, orange and yellow fluorescent proteins derived from Discosoma sp. red fluorescent protein. Nat. Biotechnol. 22: 1567-1572.

Shu, X., Shaner, N.C., Yarbrough, C.A., Tsien, R.Y., Remington, S.J. 2006. Novel chromophores and buried charges control color in mFruits. Biochemistry. 45: 9639- 9647.

Sikorski, R.S., and P. Hieter. 1989. A system of shuttle vectors and yeast host strains designed for efficient manipulation of DNA in Saccharomyces cerevisiae. Genetics. 122: 19-27.

Takai, Y., Sasaki, T., Matozaki, T. 2001. Small GTP-binding proteins. Physiol. Rev. 81(1): 153-208.

Takeshige, K., Baba, M., Tsuboi, S., Noda, T., Ohsumi, Y. 1992. Autophagy in yeast demonstrated with proteinase-deficient mutants and conditions for its induction. J. Cell. Biol. 119(2): 301-311.

Tong, A.H., Evangelista, M., Parsons, A.B., Xu, H., Bader, G.D., Pagé, N., Robinson, M., Raghibizadeh, S., Hogue, C.W., Bussey, H., Andrews, B., Tyers, M., Boone, C. Systematic genetic analysis with ordered arrays of yeast deletion mutants. Science. 294 (5550): 2364-2368.

Trotter, E.W., Grant, C.M. 2002. Thioredoxins are required for protection against a reductive stress in the yeast Saccharomyces cerevisiae. Mol. Microbiol. 46: 869-878.

Trotter, E.W., Grant, C.M. 2003. Non-reciprocal regulation of the redox state of the glutathione-glutaredoxin and thioredoxin systems. EMBO J. 4: 184-188.

Trotter, E.W., Grant, C.M. 2005. Overlapping roles of the cytoplasmic and mitochondrial redox regulatory systems in the yeast Saccharomyces cerevisiae. Eukaryotic. Cell. 4: 392- 400.

Park, H.-O., Kang, P.J., and Rachfal, A.W. (2002). Localization of the Rsr1/Bud1 GTPase involved in selection of a proper growth site in yeast. J. Biol. Chem. 277: 26721–26724. 156 Pedrajas, J.R., Miranda-Vizuete, A., Javanmardy, N., Gustafsson, J.A., Spyrou, G. 2000. Mitochondria of Saccharomyces cerevisiae contain one-conserved cysteine type peroxiredoxin with thioredoxin peroxidase activity. J. Biol. Chem. 275: 16296-16301.

Richman, T.J., Sawyer, M.M., Johnson, D.I. 2002. Saccharomyces cerevisiae Cdc42p localizes to cellular membranes and clusters at sites of polarized growth. Eukaryot Cell. 1 (3): 458-468.

Rothstein, R. 1991. Targeting, disruption, replacement, and allele rescue: Integrative DNA transformation in yeast. Methods. Enzymol. 194, 281–301.

Shatwell, K.P., Dancis, A., Cross, A.R., Klausner, R.D., Segal, A.W. 1996. The FRE1 ferric reductase of Saccharomyces cerevisiae is a cytochrome b similar to that of NADPH oxidase. J. Biol. Chem. 271(24): 14240-4.

Vanden Berghe, T., Declercq, W., Vandenabeele, P. 2007. NADPH oxidases: new players in TNF-induced necrotic cell death. Mol. Cell. 26 (6):769-771.

Veal, E. A., Day, A. M. & Morgan, B. A. 2007. Hydrogen peroxide sensing and signaling. Mol. Cell. 26: 1-14.

Vida, T. A., Emr, S. D. 1995. A new vital stain for visualizing vacuolar membrane dynamics and endocytosis in yeast. J. Cell. Biol. 128: 779-792.

Vilella, F., Herrero, E., Torres, J., de la Torre-Ruiz, M.A. 2005. Pkc1 and the upstream elements of the cell integrity pathway in Saccharomyces cerevisiae, Rom2 and Mtl1, are required for cellular responses to oxidative stress. J Biol Chem. 280 (10): 9149-59.

Wanke, V., Accorsi, K., Porro, D., Esposito, F., Russo, T., Vanoni, M. 1999. In budding yeast, reactive oxygen species induce both RAS-dependent and RAS-independent cell cycle-specific arrest. Mol. Microbiol. 32: 753-764.

Werner, E. 2004. GTPases and reactive oxygen species: switches for killing and signaling J. Cell. Sci. 117: 143-153.

Winter-Vann, A.M., Casey, P.J. 2005. Post-prenylation-processing enzymes as new targets in oncogenesis. Nat. Rev. Cancer. 5 (5): 405-12.

Winzeler, E. A. et al. 1999. Functional characterization of the S. cerevisiaegenome by gene deletion and parallel analysis. Science285, 901–906.

Wissing, S., Ludovico, P., Herker, E., Buttner, S., Engelhardt, S. M., Decker, T., Link, A., Proksch, A., Rodrigues, F., Corte-Real, M. 2004. An AIF orthologue regulates apoptosis in yeast. J. Cell. Biol. 166: 969-974.

157 Wong, H.L., Sakamoto, T., Kawasaki, T., Umemura, K., Shimamoto, K. 2004. Down- regulation of metallothionein, a reactive oxygen scavenger, by the small GTPase OsRac1 in rice. Plant. Physiol. 135: 1447-1456.

Wright, L.P., Philips, M.R. 2006. Thematic review series: lipid posttranslational modifications. CAAX modification and membrane targeting of Ras. J. Lipid. Res. 47 (5): 883-891.

Yamochi, W., Tanaka, K., Nonaka, H., Maeda, A., Musha, T., Takai, Y. 1994. Growth site localization of Rho1 small GTP-binding protein and its involvement in bud formation in Saccharomyces cerevisiae. J. Cell. Biol. 125(5): 1077-1093.

Zhang, N.N., Dudgeon, D.D., Paliwal, S., Levchenko, A., Grote, E., and Cunningham, K.W. 2006. Multiple signaling pathways regulate yeast cell death during the response to mating pheromones. Mol. Biol. Cell. 17, 3409-3422.

158